An engineering cost construction control method and system, and a storage medium

Abstract

This invention discloses a construction cost control method, system, and storage medium, relating to the field of construction cost management technology. The method includes: S1, constructing a holographic identification framework for implicit costs throughout the entire construction cycle, defining the types, occurrence scenarios, and accounting rules of various implicit costs, covering all implicit cost occurrence stages in the entire process of construction preparation, main construction, decoration and renovation, and final acceptance; This invention achieves comprehensive coverage and systematic identification of various implicit costs by constructing a holographic identification framework for implicit costs throughout the entire construction cycle, and establishes a correlation network between implicit costs and multi-dimensional influencing factors by collecting unstructured data of the construction process in real time through IoT devices and combining it with knowledge graph technology, thereby achieving accurate identification and source analysis of implicit cost inducing factors and transmission paths, and achieving early warning and proactive intervention of implicit costs by setting a dynamic threshold early warning mechanism.

Description

Technical Field

[0001] This invention relates to the field of engineering cost management technology, specifically to an engineering cost construction control method, system, and storage medium. Background Technology

[0002] Engineering refers to the construction of various buildings, structures, and their ancillary facilities, as well as the installation of supporting lines, pipelines, and equipment, to achieve specific functions. It is a crucial support for national economic development and directly relates to social infrastructure construction and the improvement of people's livelihoods. Engineering cost refers to all construction expenses estimated or actually incurred during the construction period of an engineering project, covering the entire lifecycle from project decision-making, design, construction to final acceptance. It is the core basis for investment decisions, fundraising, cost control, and benefit evaluation of engineering projects, directly determining the project's economic benefits and return on investment. Construction cost control refers to the management activities of supervising, adjusting, and controlling various expenses of an engineering project during the construction phase to ensure that the actual cost of the project does not exceed the budget target. The construction phase is the critical stage for the formation of engineering costs; cost control at this stage directly determines the final actual cost of the project and is a core link in achieving the project's investment objectives.

[0003] However, existing engineering cost control methods have certain shortcomings. Current technologies primarily focus on the accounting and management of explicit costs such as labor, materials, and machinery, while paying insufficient attention to implicit costs such as capital tied up due to project delays, idle costs caused by poor workflow coordination, waste costs from material losses, and rework costs due to quality defects. This often results in actual project costs far exceeding the budget. The lack of a systematic framework for identifying implicit costs means that these costs cannot be comprehensively covered in all scenarios. The factors inducing implicit costs are complex and their transmission paths are hidden, making it difficult for existing technologies to accurately pinpoint their causes and scope of impact. The lack of a real-time dynamic early warning mechanism means that implicit costs are often only discovered after significant losses have occurred, making early intervention impossible. Control measures are mostly reactive and lack specificity, failing to fundamentally reduce the probability of implicit costs occurring. Therefore, developing an engineering cost control method, system, and storage medium is of great significance. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method, system and storage medium for construction cost control in engineering projects. It can effectively solve the problems of lack of hidden cost control, incomplete identification, inaccurate positioning, untimely early warning and non-targeted control in the prior art, significantly improve the comprehensiveness and accuracy of construction cost control in engineering projects, reduce the actual cost of projects and ensure the smooth realization of project investment goals.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: Firstly, a method for controlling construction costs, the method comprising the following steps: Step S1: Construct a holographic identification framework for hidden costs throughout the entire construction cycle, defining the types, occurrence scenarios, and accounting rules of various hidden costs, covering all hidden cost occurrence stages throughout the entire construction preparation, main construction, decoration and renovation, and final acceptance phases. Step S2: Deploy various types of IoT sensing devices at the construction site to collect multi-dimensional unstructured data in real time, including personnel work status, machinery operating parameters, material flow information, process progress data, and environmental changes during the construction process. The collected data is then preprocessed and stored in a standardized manner. Step S3: Based on historical construction data and industry experience, construct a knowledge graph of implicit costs and establish a network of relationships between implicit costs and construction behavior, management decisions, and the external environment; Step S4: Analyze the preprocessed real-time data using deep learning algorithms, combine the implicit cost knowledge graph to identify the inducing factors and transmission paths of implicit costs, and mark potential implicit cost occurrence nodes. Step S5: Quantify and calculate the identified hidden costs according to the preset accounting rules, generate a hidden cost statistical report, and present the distribution and trend of hidden costs through a visual interface. At the same time, it supports source tracing analysis of any hidden cost item. Step S6: Set up a dynamic threshold early warning mechanism to automatically adjust the early warning threshold according to project characteristics, construction stage and cost control objectives. When the cumulative value of a certain type of hidden cost approaches or exceeds the corresponding threshold, the corresponding level of early warning signal is triggered. Step S7: Based on the early warning information and the results of the source analysis, combined with the actual construction situation, generate targeted hidden cost control measures and automatically push them to relevant management personnel. At the same time, track the implementation effect of the control measures and continuously optimize the control strategy.

[0006] Furthermore, step S2 includes the following steps when preprocessing the collected data: The raw collected data is cleaned to remove duplicate and invalid data, and the collected data of different formats is converted to a standard data format that the system can recognize. Outliers in the data are identified and processed, missing data are filled in using interpolation, and data of different dimensions are normalized to eliminate the impact of dimensional differences on subsequent analysis. Timestamp alignment is performed on all processed data to ensure consistency across different sources in terms of time.

[0007] Furthermore, step S3, in constructing the implicit cost knowledge graph, includes the following steps: Extract entities and relationships related to implicit costs from historical construction data and industry literature, and construct entity and relationship databases; The extracted entities and relations are standardized, entity naming and relation definition are unified, the ontology structure of the knowledge graph is constructed based on the entities and relations, and the attributes of various entities and the constraints of relations are defined. The processed entities and relationships are imported into the ontology structure to generate an initial implicit cost knowledge graph; The initial knowledge graph is supplemented and corrected through manual review and machine learning algorithms to form the final implicit cost knowledge graph.

[0008] Furthermore, step S4, when analyzing the preprocessed real-time data using a deep learning algorithm, includes the following steps: The pre-processed real-time data is input into a pre-trained deep learning model to extract feature information from the data. The extracted feature information is matched with entities and relationships in the implicit cost knowledge graph to determine the type of implicit cost related to the current data; Based on the network of relationships in knowledge graphs, we analyze the inducing factors and transmission paths of implicit costs; Based on the analysis results, potential hidden cost occurrence nodes are marked, and the risk level of each node is recorded. The node risk level is calculated using the following formula: ,in, Let i be the risk level of the i-th node where implicit costs occur. Let be the weight of the j-th transmission path corresponding to the i-th node. This weight is obtained by statistically analyzing the frequency of implicit costs actually occurring due to this transmission path in historical construction data. This is the real-time feature matching degree coefficient for the j-th transmission path corresponding to the i-th node. This coefficient is calculated by weighting the importance parameters of each feature dimension automatically learned during the training of the deep learning model.

[0009] Furthermore, the accounting rules for various implicit costs defined in step S1 include rules for capital occupation cost accounting, idle labor cost accounting, material loss cost accounting, and rework cost accounting. The capital occupation cost accounting rule is calculated based on the project capital cost rate and the duration of capital occupation. The idle labor cost accounting rule is calculated based on the number of idle workers, the duration of idle work, and the daily wage standard for workers. The material loss cost accounting rule is calculated based on the difference between the actual consumption of materials and the quota consumption. The rework cost accounting rule is calculated based on the amount of rework and the unit cost of the work.

[0010] Furthermore, in step S5, when performing source tracing analysis on any implicit cost item, starting from the node where the implicit cost occurs, the process traces backward along the transmission path in the knowledge graph, sequentially acquiring relevant data and information for each transmission node, including the time of occurrence, personnel involved, machinery used, materials consumed, processes performed, and management decisions made. Through comprehensive analysis of the above data and information, the root cause of the implicit cost is determined, and a source tracing analysis report is generated. The contribution of each root cause node is calculated using the following formula: ,in, Let k be the contribution of the k-th root cause node. Let be the path length from the k-th root cause node to the node where the hidden cost occurs. Let be the intrinsic influence strength coefficient of the k-th root cause node. This coefficient was determined by industry experts based on statistical analysis of numerous engineering cases. is the real-time impact coefficient of the k-th root cause node, which is quantified by the degree of abnormality of the node under the current construction state.

[0011] Furthermore, in step S6, when automatically adjusting the early warning threshold based on project characteristics, construction stage, and cost control objectives, the basic information of the project's investment scale, engineering type, construction difficulty, and contract period is first obtained. Then, the cost control weight of the current construction stage is determined. Next, the early warning thresholds for various implicit costs in this stage are calculated in conjunction with the overall cost control objectives of the project. Finally, the early warning thresholds are dynamically adjusted in real time based on the actual cost consumption during the construction process. The dynamic early warning threshold is calculated using the following formula: ,in, Let be the dynamic early warning threshold for the m-th type of implicit cost. Let be the initial baseline threshold for the m-th type of implicit cost, which is determined by the average implicit cost data of completed projects of the same type. This is the cost control weighting coefficient for the current construction phase. This coefficient is determined by the cost allocation ratio for each phase in the overall project cost control plan. This is the historical cost deviation correction factor, which is obtained by statistically analyzing the actual deviation rate of similar hidden costs during the early stages of construction in this project. This is the correction factor for the current progress completion rate, which is determined by the ratio of the current completed work volume to the total work volume.

[0012] Furthermore, in step S7, when tracking the implementation effect of control measures, relevant construction data after the implementation of control measures are collected periodically. The implicit cost data after implementation is compared with the implicit cost data before implementation to analyze the degree of impact of control measures on implicit costs. Based on the analysis results, control measures are adjusted and optimized. At the same time, the implementation status and effect data of control measures are stored in the historical database to provide a reference for cost control of subsequent projects.

[0013] Secondly, an engineering cost construction control system is provided, applicable to the aforementioned engineering cost construction control method. The system includes: an identification framework construction module, a data acquisition module, a knowledge graph construction module, an analysis and identification module, an accounting visualization module, an early warning module, and a control measure generation module. The identification framework construction module is used to build a holographic identification framework for hidden costs throughout the construction cycle, defining the types, occurrence scenarios, and accounting rules of various hidden costs; The data acquisition module is connected to multiple IoT sensing devices to collect multi-dimensional unstructured data in real time during the construction process, and to preprocess and standardize the data storage. The knowledge graph construction module is used to construct a hidden cost knowledge graph and establish a network of relationships between hidden costs and construction behavior, management decisions, and the external environment. The analysis and identification module is used to analyze the preprocessed real-time data using deep learning algorithms, and combined with the implicit cost knowledge graph, to identify the inducing factors and transmission paths of implicit costs, and mark potential implicit cost occurrence nodes. The accounting visualization module is used to quantify and calculate the identified hidden costs, generate statistical reports and present them visually, and at the same time realize the source analysis of hidden costs. The early warning module is used to set a dynamic threshold early warning mechanism, which automatically adjusts the early warning threshold according to project characteristics, construction stage and cost control objectives. When the cumulative value of a certain type of hidden cost approaches or exceeds the corresponding threshold, the corresponding level of early warning signal is triggered. The control measures generation module is used to generate targeted hidden cost control measures based on early warning information and source analysis results, push them to relevant management personnel, track the implementation effect, and continuously optimize the control strategy.

[0014] Thirdly, a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method for controlling construction costs.

[0015] Compared with existing technologies, the construction cost control method, system, and storage medium of this project have the following advantages: This invention constructs a holographic identification framework for hidden costs throughout the entire construction cycle, achieving comprehensive coverage and systematic identification of various hidden costs. By collecting unstructured data from the construction process in real time using IoT devices and combining it with knowledge graph technology, a network of relationships between hidden costs and multi-dimensional influencing factors is established. This enables accurate identification and source analysis of hidden cost inducing factors and transmission paths. A dynamic threshold early warning mechanism is set up to provide early warning and proactive intervention for hidden costs. By generating targeted control measures and tracking their implementation effects, precise control and continuous optimization of hidden costs are achieved. This invention effectively solves the problems of missing hidden cost control, incomplete identification, inaccurate positioning, untimely early warning, and untargeted control in existing technologies. It significantly improves the comprehensiveness and accuracy of construction cost control, reduces actual project costs, and ensures the smooth achievement of project investment goals.

[0016] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0018] Figure 1 A flowchart of a construction cost control method for engineering projects; Figure 2 A flowchart illustrating the steps of a construction cost control method for engineering projects; Figure 3 This is a schematic diagram of a construction cost control system for engineering projects. Detailed Implementation

[0019] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0020] The core of this invention is a system and computer-readable storage medium for a method of controlling hidden costs throughout the entire construction cycle of engineering projects. This solution is based on the holographic identification of hidden costs throughout the construction cycle. It collects multi-dimensional unstructured construction data through IoT devices at the construction site, relies on a hidden cost knowledge graph and deep learning algorithms to accurately identify the inducing factors and transmission paths of hidden costs, completes the quantitative calculation and visualization of hidden costs according to preset accounting rules, and combines a dynamic threshold early warning mechanism to achieve early warning of hidden cost risks. Finally, it generates targeted control measures and tracks and optimizes their implementation. At the system level, the above method flow is realized through the collaborative operation of modules including an identification framework construction module, a data acquisition module, a knowledge graph construction module, an analysis and identification module, an accounting and visualization module, an early warning module, and a control measure generation module. The computer-readable storage medium stores executable programs to programmatically implement the method steps. The overall technical solution achieves comprehensive identification, accurate tracing, dynamic early warning, and proactive control of hidden costs during the construction phase of engineering projects, overcoming the shortcomings of traditional cost control methods that only focus on explicit costs.

[0021] This embodiment uses a high-rise residential building project as a practical application. The project covers four complete stages: construction preparation, main construction, decoration, and final acceptance. It fully implements the engineering cost and construction control method and supporting system of this patented invention, achieving refined and intelligent management of hidden costs throughout the entire construction cycle. Figure 1 The diagram shown is an overall flowchart of the construction cost control method in this embodiment. Figure 2 The diagram shown is a flowchart of the method steps, as follows: Figure 3 The diagram shown is a schematic of the control system structure. The specific implementation process of this embodiment will be described in detail below with reference to the attached diagram and actual construction scenario.

[0022] This embodiment first activates the identification framework construction module to build a holographic identification framework for implicit costs covering the entire construction process. This framework comprehensively covers all implicit cost occurrence stages in the four phases of construction preparation, main construction, decoration, finishing, and acceptance, clearly defining the specific types, scenarios, and standardized accounting rules for various implicit costs. This embodiment focuses on defining four core implicit costs: capital occupation cost, idle labor cost, material loss cost, and rework cost, and develops specific accounting rules for each type. The capital occupation cost accounting rule uses the project capital cost rate and capital occupation duration as the core calculation basis to accurately calculate the implicit costs caused by construction delays and poor capital turnover. The idle labor cost accounting rule uses the number of idle workers, idle time, and daily wage standards as the calculation basis to quantify the labor idle costs caused by poor process coordination and personnel scheduling errors. The material loss cost accounting rule uses the difference between the actual material consumption and the quota consumption as the calculation basis to statistically analyze the material waste costs caused by improper on-site management and material storage errors. The rework cost accounting rules are based on the amount of rework and the unit cost of the work, calculating the rework losses caused by defects in construction quality and substandard processes. This framework transforms implicit costs from qualitative descriptions to quantitative accounting, laying the foundation for subsequent full-process management.

[0023] This embodiment relies on a data acquisition module to deploy various types of IoT sensing devices at the construction site of a high-rise residential building. These devices include personnel positioning terminals, machinery operation sensors, material RFID identification equipment, process progress collectors, and environmental monitoring instruments, enabling real-time collection of multi-dimensional unstructured data throughout the construction process. The collected data includes five main categories: personnel work status, machinery operation parameters, material flow information, process progress data, and environmental changes, comprehensively covering dynamic information from all stages of construction production.

[0024] Immediately after data acquisition, the preprocessing stage begins. First, the raw data undergoes data cleaning to remove duplicates and invalid or outlier data, and heterogeneous data from different devices is converted into a standardized data format recognizable by the system. Next, outliers are identified and processed, and interpolation is used to accurately fill in missing values. Data with different units of measurement is normalized to eliminate interference from dimensional differences in subsequent data analysis. Finally, all preprocessed data undergoes timestamp alignment to ensure high consistency across different sources such as personnel, machinery, materials, processes, and environment, providing standardized data support for subsequent deep learning analysis and knowledge graph matching.

[0025] This embodiment utilizes a knowledge graph construction module to build a dedicated knowledge graph for implicit costs based on historical construction data and general experience data from the construction industry. First, it extracts entities and relationships related to implicit costs from historical construction databases and authoritative industry literature, constructing standardized entity and relationship libraries. Entities cover core elements such as implicit cost types, construction behavior, management decisions, and the external environment; relationships cover logical links such as induction, cause, and associated impact. Second, the extracted entities and relationships are standardized, unifying entity naming rules and relationship definition standards. Based on these standardized entities and relationships, a knowledge graph ontology structure is built, clarifying the attribute characteristics of various entities and the constraints of relationships. Subsequently, the processed entities and relationships are imported into the ontology structure, generating an initial version of the implicit cost knowledge graph. Finally, combining manual review by industry experts and automatic optimization using machine learning algorithms, the initial knowledge graph is completed and corrected, eliminating erroneous associations and supplementing missing elements, forming a final implicit cost knowledge graph directly applicable to this project. This graph fully establishes a network of relationships between implicit costs and the external environment of construction behavior management decisions, achieving a visualized and structured presentation of the implicit cost transmission logic.

[0026] This embodiment initiates the analysis and identification module, inputting pre-processed standardized real-time data into a pre-trained deep learning model. The model extracts core feature information from the data and then precisely matches this extracted feature information with entity relationships in the implicit cost knowledge graph to quickly determine the type of implicit cost corresponding to the current construction data. Based on the relational network in the knowledge graph, it deeply analyzes the inducing factors and transmission paths of implicit costs, accurately marks potential implicit cost occurrence nodes during construction, and quantifies the risk level of each node.

[0027] In the specific implementation of this embodiment, the node risk level is calculated using the following formula: ,in Let i be the risk level of the i-th node where implicit costs occur. The weight of the j-th transmission path corresponding to the i-th node is obtained by statistically analyzing the frequency of the implicit costs actually occurring due to this transmission path in historical construction data. The higher the frequency of occurrence, the greater the weight. This is the real-time feature matching degree coefficient for the j-th transmission path corresponding to the i-th node. This coefficient is obtained by automatically learning the importance parameters of each feature dimension and weighting them during the deep learning model training process. The higher the feature matching degree, the larger the coefficient value. This formula quantifies the risk level of all nodes where potential hidden costs occur, providing accurate data for subsequent early warning and control.

[0028] This embodiment utilizes an accounting visualization module to quantify and calculate identified hidden costs according to the accounting rules within the previously constructed holographic recognition framework. It automatically generates hidden cost statistical reports and presents the phased distribution and real-time trends of various hidden costs through a visual interface, facilitating managers' rapid understanding of cost dynamics. Simultaneously, it performs source tracing analysis on any hidden cost item. Starting from the point where the hidden cost occurs, it traces backward along the transmission path in the knowledge graph, sequentially obtaining complete information such as the occurrence time of the transmission node, personnel usage, machinery consumption, material consumption, process management decisions, etc. This comprehensive analysis determines the root cause of the hidden cost and generates a professional source tracing analysis report.

[0029] In the specific implementation of this embodiment, the contribution of each root cause node is calculated using the following formula: ,in Let k be the contribution of the k-th root cause node. The length of the transmission path from the k-th root cause node to the node where the hidden cost occurs is denoted as . The longer the path length, the smaller the contribution value. This is the inherent influence strength coefficient of the k-th root cause node. This coefficient is determined by industry experts based on statistical analysis results of a large number of similar engineering cases. The stronger the inherent influence of the root cause node, the larger the coefficient value is assigned. This represents the real-time impact coefficient of the k-th root cause node. This coefficient is quantified by the degree of anomaly of the node under the current construction status; the higher the degree of anomaly, the larger the coefficient. This formula accurately calculates the contribution of each root cause node, identifies the core causes of hidden costs, and provides direction for the formulation of control measures.

[0030] This embodiment relies on an early warning module to build a dynamic threshold early warning mechanism. First, it obtains basic information such as the investment scale, project type, construction difficulty, and contract period of the high-rise residential project. Then, it determines the cost control weight of the current construction stage, calculates the initial early warning threshold for various implicit costs in the current stage in combination with the overall cost control target of the project, and then dynamically adjusts the early warning threshold in real time according to the actual cost consumption during the construction process.

[0031] In the specific implementation of this embodiment, the dynamic early warning threshold is calculated using the following formula: ,in, Let be the dynamic early warning threshold for the m-th type of implicit cost. The initial benchmark threshold for the m-th type of implicit cost is determined by statistical analysis of the average implicit cost data of completed projects of the same type. This is the cost control weighting coefficient for the current construction phase, which is determined by calculating the cost allocation ratio for each phase in the overall project cost control plan. This is the historical cost deviation correction factor, which is obtained by statistically analyzing the actual deviation rate of similar hidden costs during the early construction phase of this project. This is a correction factor for the current progress completion rate, calculated by the ratio of the currently completed work volume to the total work volume. The early warning mechanism monitors the cumulative values ​​of various hidden costs in real time. When the value approaches or exceeds the corresponding dynamic early warning threshold, it immediately triggers the corresponding level of early warning signal, achieving early warning of hidden cost risks.

[0032] This embodiment utilizes a control measure generation module, combining early warning information and source tracing analysis results, to generate targeted and implementable hidden cost control measures tailored to the actual construction situation of this high-rise residential project. These measures are automatically pushed to management personnel at all stages of the project. During the implementation of these control measures, post-implementation construction data and hidden cost data are collected periodically. The post-implementation data is compared and analyzed with the pre-implementation data to accurately assess the control effectiveness of the measures on hidden costs. Based on the analysis results, the control strategy is adjusted and optimized in a timely manner to ensure that the control measures continuously adapt to changes in the construction site. Simultaneously, the implementation process and effectiveness data of the control measures are completely stored in the project's historical database, providing a true and effective reference for cost control in subsequent similar projects.

[0033] During the operation of this embodiment, if Figure 3 The control system shown has modules that work together in a coordinated manner. The identification framework construction module provides the rule foundation, the data acquisition module ensures data supply, the knowledge graph construction module builds the logical network, the analysis and identification module completes intelligent analysis, the accounting and visualization module realizes data presentation, the early warning module completes risk warning, and the control measures generation module implements closed-loop control. All modules work together to achieve the complete implementation of the methodology and process.

[0034] In summary, this embodiment, through the complete implementation of the engineering cost construction control method and supporting system of this invention patent, achieves holographic identification, real-time collection, precise analysis, quantitative accounting, dynamic early warning, and proactive management of hidden costs throughout the entire construction cycle of high-rise residential buildings. It leverages IoT data collection technology to achieve comprehensive perception of construction information, utilizes knowledge graphs and deep learning algorithms to accurately trace the causes and transmission paths of hidden costs, achieves early intervention for hidden cost risks through dynamic threshold early warning, and continuously optimizes management strategies through quantitative accounting and effect tracking. This embodiment effectively addresses the industry pain points of traditional engineering cost construction control, such as incomplete identification, inaccurate positioning, untimely early warning, and lack of targeted management of hidden costs. It significantly improves the comprehensiveness and accuracy of engineering cost construction control, effectively reduces actual project construction costs, ensures the smooth achievement of project investment goals, and provides replicable and scalable standardized technical solutions and implementation experience for cost construction control of various building engineering projects, promoting the development of the engineering cost management industry towards intelligence and refinement.

[0035] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for controlling construction costs, characterized in that, The method includes the following steps: Step S1: Construct a holographic identification framework for hidden costs throughout the entire construction cycle, defining the types, occurrence scenarios, and accounting rules of various hidden costs, covering all hidden cost occurrence stages throughout the entire construction preparation, main construction, decoration and renovation, and final acceptance phases. Step S2: Deploy various types of IoT sensing devices at the construction site to collect multi-dimensional unstructured data in real time, including personnel work status, machinery operating parameters, material flow information, process progress data, and environmental changes during the construction process. The collected data is then preprocessed and stored in a standardized manner. Step S3: Based on historical construction data and industry experience, construct a knowledge graph of hidden costs and establish a network of relationships between hidden costs and construction behavior, management decisions, and the external environment; Step S4: Analyze the preprocessed real-time data using deep learning algorithms, combine the implicit cost knowledge graph to identify the inducing factors and transmission paths of implicit costs, and mark potential implicit cost occurrence nodes. Step S5: Quantify and calculate the identified hidden costs according to the preset accounting rules, generate a hidden cost statistical report, and present the distribution and trend of hidden costs through a visual interface. At the same time, it supports source tracing analysis of any hidden cost item. Step S6: Set up a dynamic threshold early warning mechanism to automatically adjust the early warning threshold according to project characteristics, construction stage and cost control objectives. When the cumulative value of a certain type of hidden cost approaches or exceeds the corresponding threshold, the corresponding level of early warning signal is triggered. Step S7: Based on the early warning information and the results of the source analysis, combined with the actual construction situation, generate targeted hidden cost control measures and automatically push them to relevant management personnel. At the same time, track the implementation effect of the control measures and continuously optimize the control strategy.

2. The method for controlling construction costs according to claim 1, characterized in that, Step S2 includes the following steps when preprocessing the collected data: The raw collected data is cleaned to remove duplicate and invalid data, and the collected data of different formats is converted to a standard data format that the system can recognize. Outliers in the data are identified and processed, missing data are filled in using interpolation, and data of different dimensions are normalized to eliminate the impact of dimensional differences on subsequent analysis. All processed data is timestamped to ensure consistency in time across different sources.

3. The method for controlling construction costs according to claim 1, characterized in that, Step S3, when constructing the implicit cost knowledge graph, includes the following steps: Extract entities and relationships related to implicit costs from historical construction data and industry literature, and construct entity and relationship databases; The extracted entities and relations are standardized, entity naming and relation definition are unified, the ontology structure of the knowledge graph is constructed based on the entities and relations, and the attributes of various entities and the constraints of relations are defined. The processed entities and relationships are imported into the ontology structure to generate an initial implicit cost knowledge graph; The initial knowledge graph is supplemented and corrected through manual review and machine learning algorithms to form the final implicit cost knowledge graph.

4. The method for controlling construction costs according to claim 1, characterized in that, Step S4, which analyzes the preprocessed real-time data using a deep learning algorithm, includes the following steps: The pre-processed real-time data is input into a pre-trained deep learning model to extract feature information from the data. The extracted feature information is matched with entities and relationships in the implicit cost knowledge graph to determine the type of implicit cost related to the current data; Based on the network of relationships in knowledge graphs, we analyze the inducing factors and transmission paths of implicit costs; Based on the analysis results, potential hidden cost occurrence nodes are marked, and the risk level of each node is recorded. The node risk level is calculated using the following formula: ,in, Let i be the risk level of the i-th node where implicit costs occur. Let be the weight of the j-th transmission path corresponding to the i-th node. is the real-time feature matching degree coefficient corresponding to the j-th transmission path of the i-th node.

5. The method for controlling construction costs according to claim 1, characterized in that, The accounting rules for various implicit costs defined in step S1 include rules for capital occupation cost accounting, idle labor cost accounting, material loss cost accounting, and rework cost accounting. Among them, the capital occupation cost accounting rule is calculated based on the project capital cost rate and the duration of capital occupation; the idle labor cost accounting rule is calculated based on the number of idle workers, the duration of idle labor, and the daily wage standard of workers; the material loss cost accounting rule is calculated based on the difference between the actual consumption of materials and the quota consumption; and the rework cost accounting rule is calculated based on the amount of rework and the unit cost of the work.

6. The method for controlling construction costs according to claim 1, characterized in that, In step S5, when performing source tracing analysis on any implicit cost item, starting from the node where the implicit cost occurs, the process traces backward along the transmission path in the knowledge graph, sequentially acquiring relevant data and information for each transmission node, including the time of occurrence, personnel involved, machinery used, materials consumed, procedures performed, and management decisions made. Through comprehensive analysis of the above data and information, the root cause of the implicit cost is determined, and a source tracing analysis report is generated. The contribution of each root cause node is calculated using the following formula: ,in, Let k be the contribution of the k-th root cause node. Let be the path length from the k-th root cause node to the node where the hidden cost occurs. Let be the intrinsic influence strength coefficient of the k-th root cause node. is the real-time impact coefficient of the k-th root cause node.

7. The method for controlling construction costs according to claim 1, characterized in that, In step S6, when automatically adjusting the early warning threshold based on project characteristics, construction stage, and cost control objectives, the basic information of the project's investment scale, engineering type, construction difficulty, and contract period is first obtained. Then, the cost control weight of the current construction stage is determined. Next, the early warning thresholds for various implicit costs in this stage are calculated in conjunction with the overall cost control objectives of the project. Finally, the early warning thresholds are dynamically adjusted in real time based on the actual cost consumption during construction. The dynamic early warning threshold is calculated using the following formula: ,in, Let be the dynamic early warning threshold for the m-th type of implicit cost. Let be the initial baseline threshold for the m-th type of implicit cost. This represents the cost control weighting coefficient for the current construction phase. This is the historical cost deviation correction factor. This is a correction factor for the current progress completion rate.

8. The method for controlling construction costs according to claim 1, characterized in that, In step S7, when tracking the implementation effect of control measures, relevant construction data after the implementation of control measures are collected regularly. The implicit cost data after implementation is compared with the implicit cost data before implementation to analyze the degree of impact of control measures on implicit costs. Based on the analysis results, control measures are adjusted and optimized. At the same time, the implementation status and effect data of control measures are stored in the historical database to provide a reference for cost control of subsequent projects.

9. A construction cost control system, applicable to the construction cost control method described in any one of claims 1-8, characterized in that, The system includes: an identification framework construction module, a data acquisition module, a knowledge graph construction module, an analysis and identification module, an accounting and visualization module, an early warning module, and a control measure generation module; The identification framework construction module is used to build a holographic identification framework for hidden costs throughout the construction cycle, defining the types, occurrence scenarios, and accounting rules of various hidden costs; The data acquisition module is connected to multiple IoT sensing devices to collect multi-dimensional unstructured data in real time during the construction process, and to preprocess and standardize the data storage. The knowledge graph construction module is used to construct a hidden cost knowledge graph and establish a network of relationships between hidden costs and construction behavior, management decisions, and the external environment. The analysis and identification module is used to analyze the preprocessed real-time data using deep learning algorithms, and combined with the implicit cost knowledge graph, to identify the inducing factors and transmission paths of implicit costs, and mark potential implicit cost occurrence nodes. The accounting visualization module is used to quantify and calculate the identified hidden costs, generate statistical reports and present them visually, and at the same time realize the source analysis of hidden costs. The early warning module is used to set a dynamic threshold early warning mechanism, which automatically adjusts the early warning threshold according to project characteristics, construction stage and cost control objectives. When the cumulative value of a certain type of hidden cost approaches or exceeds the corresponding threshold, the corresponding level of early warning signal is triggered. The control measures generation module is used to generate targeted hidden cost control measures based on early warning information and source analysis results, push them to relevant management personnel, track the implementation effect, and continuously optimize the control strategy.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the construction cost control method according to any one of claims 1-8.

Images