Self-adaptive construction engineering cost control method and system

Abstract

The invention belongs to the technical field of constructional engineering, and particularly relates to a self-adaptive constructional engineering cost control method and system. A load sensor installed on a tower crane is used for monitoring the weight of a hoisted material, the number of hoisting times is combined, the real-time using amount of the material is calculated through a formula, and a big data analysis technology and a machine learning algorithm are used for constructing a cost prediction model. Factors such as manual efficiency change coefficients at different construction stages, shutdown time length caused by abnormal weather and additional cost coefficients caused by the shutdown time length are considered, and the cost change trend can be accurately reflected; the problems that a traditional construction engineering cost control method cannot adapt to various uncertain factors in engineering, so that cost prediction is not accurate, and coping strategies are lacked are solved, cost hyper-branched and resource waste are prevented, and the problems of accurate control and efficient management of cost are solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of construction engineering, and in particular relates to an adaptive construction engineering cost control method and system. Background Art

[0002] In the construction sector, traditional cost control methods have exposed numerous flaws in practical application. They often rely on static budgets and fixed standards for monitoring and adjustment, making them unable to cope with the complexity and dynamic nature of construction projects. Design changes frequently occur during construction. For example, adjustments to building functional requirements lead to structural layout changes, resulting in significant discrepancies between the original budgeted material usage and actual usage. Raw material market prices fluctuate drastically, with steel and cement prices fluctuating significantly due to supply and demand and macroeconomic factors, often leading to imbalanced cost budgets. Construction conditions are complex and ever-changing, with poor geological conditions increasing foundation construction costs and severe weather leading to work stoppages, delays, and additional costs. These uncertainties make it difficult for traditional methods to accurately control costs, leading to frequent overruns and resource waste, seriously impacting project economic benefits and management effectiveness. Summary of the Invention

[0003] In response to the above-mentioned problems existing in the prior art, the purpose of the present invention is to provide an adaptive construction project cost control method, which can solve the problems of inaccurate cost forecasts and lack of response strategies caused by the inability of traditional construction project cost control methods to adapt to various uncertainties in the project, thereby preventing cost overruns and waste of resources, and solving the problems of accurate cost control and efficient management.

[0004] In order to solve the above problems, the technical solutions adopted by the present invention are as follows: An adaptive construction project cost control method, the method comprising the following steps: S1: Monitor the weight of lifted materials , combined with the number of lifting times , calculate the real-time weight of the material by the following formula: ; in, The weight of the material lifted each time; Collect weather conditions data and real-time price data of raw materials , set up a manual input port to collect time-consuming data of special construction processes ; S2: Introduce big data analysis to deeply process the collected data and build a system based on the project progress. The horizontal axis shows cost changes. The vertical axis is the cost prediction model, which takes into account the change coefficient of labor efficiency in different construction stages. , the length of downtime caused by abnormal weather and the additional cost coefficient it causes , set up a cost prediction model and continuously optimize the model through machine learning algorithms so that it can dynamically calibrate the prediction results based on real-time inflow data; S3: Based on the cost prediction model set in step S2, the project cost is monitored in real time. Combining the project characteristics with the cost budget planning, independent thresholds are set for the material cost and labor cost of the preset key construction nodes. , labor costs ; When the material cost forecast value of a preset key construction stage is If the upward trend meets the preset formula, or the labor cost deviates from the normal pickup rate by a preset ratio, an early warning will be triggered: S4: At the moment the warning is triggered, the system selects higher-quality alternative material suppliers based on the built-in knowledge base and algorithm, adjusts the construction sequence to prioritize processes that are relatively less affected by price fluctuations, and then allocates existing inventory materials based on the process cost sensitivity function, providing an estimate of the effects of each strategy after implementation.

[0005] Furthermore, in step S2, the cost prediction model is: ; in, Indicates different construction stages.

[0006] Furthermore, in step S3, the preset formula is: ; in, for The predicted value of material cost at the moment, for The forecast value of material cost at the moment.

[0007] Furthermore, in step S4, the sensitivity function is: ; in, For the A process, is the total cost.

[0008] An adaptive construction project cost control system, the system includes a data acquisition module, a data analysis and modeling module, a monitoring and early warning module, and a strategy generation module; The data acquisition module includes a sensor unit, a data interface unit and a manual input unit; The sensor unit includes a tower crane load sensor, an environmental sensor, and an equipment operation status sensor. The sensor unit is used to convert physical quantities at the construction site into electrical signals or digital signals, and capture preset key data parameters in real time; The data interface unit is used to establish a dedicated data interface with various supplier systems including those for building materials and equipment rental; The manual input unit uses a human-computer interaction interface to allow frontline construction workers to input working condition information into the system via a mobile terminal or a fixed terminal; The data analysis and modeling module includes a data processing unit and a model building and optimization unit; The data processing unit has built-in data cleaning, conversion and aggregation algorithms to pre-process the heterogeneous data in the system, remove outliers, fill in missing values, and regularize the processed data into a unified standard format; The model building optimization unit integrates a machine learning algorithm library, automatically screens suitable algorithms based on engineering characteristics and builds an initial cost prediction model. At the same time, the model building optimization unit continuously adjusts model parameters based on real-time feedback data; The monitoring and early warning module includes a real-time comparison unit and an early warning push unit; The real-time comparison unit is configured to continuously obtain actual cost data and predicted cost data output by the data analysis and modeling module, and compare the differences to locate the source of cost fluctuations; The warning push unit pushes alarm information to project managers at all levels according to preset warning rules when the cost deviation reaches a threshold; The strategy generation module includes a knowledge base unit and an intelligent decision-making unit; The knowledge base unit is configured to store historical project cost control cases, including information such as project type, project scale, success records and failure records, as well as implementation steps, resource inputs and effectiveness feedback data of historical cost control strategies; The intelligent decision-making unit combines the current real-time status of the project and the reserve data of the knowledge base unit to generate a customized cost control strategy combination through a preset algorithm.

[0009] Furthermore, in the model construction optimization unit, a project progress The horizontal axis shows cost changes. The vertical axis is the cost prediction model, which takes into account the change coefficient of labor efficiency in different construction stages. , the length of downtime caused by abnormal weather and the additional cost coefficient it causes , the cost prediction model is: ; in, Indicates different construction stages.

[0010] Furthermore, in the intelligent decision-making unit, the strategy generation basis is the cost optimization objective function and the process cost sensitivity function, and the sensitivity function is: ; in, For the A process, is the total cost.

[0011] Compared with the prior art, the present invention has the following beneficial effects: (1) This invention uses big data analysis technology and machine learning algorithms to build a cost prediction model, taking into account factors such as the labor efficiency variation coefficient at different construction stages, the downtime caused by abnormal weather and the additional cost coefficient caused by it. It can accurately reflect the cost change trend and overcome the problem of inaccurate cost prediction caused by traditional methods due to information limitations and lack of dynamic analysis; (2) The present invention monitors the project cost in real time based on the constructed model. Combining the project characteristics with the cost budget planning, the present invention sets independent thresholds for the material cost and labor cost of key construction nodes. The thresholds can be dynamically adjusted according to the actual situation of the project. When the cost forecast value fluctuates abnormally, such as the material cost of a key construction stage rises by more than 10% in a short period of time or the labor cost deviates from the normal range by a certain proportion, it can immediately trigger a precise early warning, timely discover the cost risk, and change the passive situation of post-adjustment of the traditional method. (3) The system of the present invention has a built-in massive knowledge base and intelligent algorithms. At the moment of early warning triggering, it can quickly generate multiple sets of targeted strategies based on the experience of coping with similar engineering cases in the past and the evaluation data of the implementation of various cost control strategies. For example, when the price of raw materials rises sharply, it can recommend the urgent expansion and screening of high-quality alternative material suppliers, cleverly adjust the construction sequence to give priority to the processes that are less affected by price fluctuations, and reasonably allocate existing inventory materials according to the process cost sensitivity function to delay high-priced purchases, etc. It also provides an estimate of the effect of each strategy after implementation, assists managers in making efficient decisions, and effectively solves the dilemma of traditional methods lacking coping strategies when cost problems arise, improves the level of cost management of construction projects, avoids cost overruns and waste of resources, and ensures the economic benefits of the project. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 This is a schematic diagram of the module structure of the control system described in the present invention. DETAILED DESCRIPTION

[0013] The present invention will be further described below with reference to specific embodiments.

[0014] The adaptive construction project cost control method of the present invention comprises the following steps: S1: Monitor the weight of lifted materials , combined with the number of lifting times , calculate the real-time weight of the material by the following formula: ; in, The weight of the material lifted each time; Collect weather conditions data and real-time price data of raw materials , set up a manual input port to collect time-consuming data of special construction processes ; S2: Introduce big data analysis to deeply process the collected data and build a system based on the project progress. The horizontal axis shows cost changes. The vertical axis is the cost prediction model, which takes into account the change coefficient of labor efficiency in different construction stages. , the length of downtime caused by abnormal weather and the additional cost coefficient it causes , set up a cost prediction model and continuously optimize the model through machine learning algorithms so that it can dynamically calibrate the prediction results based on real-time inflow data; The above cost prediction model is: ; in, Indicates different stages of construction; S3: Based on the cost prediction model set in step S2, the project cost is monitored in real time. Combining the project characteristics with the cost budget planning, independent thresholds are set for the material cost and labor cost of the preset key construction nodes. , labor costs ; When the material cost forecast value of a preset key construction stage is If the upward trend meets the preset formula, or the labor cost deviates from the normal pickup rate by a preset ratio, an early warning will be triggered: The above preset formula is: ; in, for The predicted value of material cost at the moment, for The forecast value of material cost at the moment; S4: When an alert is triggered, the system uses its built-in knowledge base and algorithms to identify suppliers of higher-quality alternative materials. It then adjusts the construction sequence to prioritize processes that are less affected by price fluctuations. It then allocates existing inventory materials based on a process cost sensitivity function and provides an estimate of the effectiveness of each strategy after implementation. The sensitivity function mentioned above is: ; in, For the A process, is the total cost.

[0015] The adaptive construction project cost control system described in the present invention further includes a data acquisition module, a data analysis and modeling module, a monitoring and early warning module, and a strategy generation module; The data acquisition module includes a sensor unit, a data interface unit and a manual input unit; The sensor unit includes tower crane load sensors, environmental sensors, and equipment operation status sensors. The sensor unit is used to convert physical quantities on the construction site into electrical or digital signals, capturing preset key data parameters in real time. The data interface unit is used to build dedicated data interfaces with various supplier systems, including those for building materials and equipment rental; The manual input unit uses a human-computer interaction interface to allow frontline construction workers to enter working condition information into the system via mobile or fixed terminals; The data analysis and modeling module includes a data processing unit and a model building and optimization unit; The data processing unit has built-in data cleaning, conversion, and aggregation algorithms to pre-process heterogeneous data in the system, remove outliers, fill in missing values, and normalize the processed data into a unified standard format; The model building and optimization unit integrates a machine learning algorithm library, automatically screening suitable algorithms based on engineering characteristics and building an initial cost prediction model. At the same time, the model building and optimization unit continuously adjusts model parameters based on real-time feedback data. The monitoring and early warning module includes a real-time comparison unit and an early warning push unit; The real-time comparison unit is configured to continuously obtain actual cost data and the predicted cost data output by the data analysis and modeling module, compare the differences, and locate the source of cost fluctuations; The early warning push unit pushes alarm information to project managers at all levels according to the preset early warning rules when the cost deviation reaches the threshold; The strategy generation module includes a knowledge base unit and an intelligent decision-making unit; The knowledge base unit is configured to store historical project cost control cases, covering information such as project type, project scale, success records and failure records, as well as the implementation steps, resource input and effectiveness feedback data of historical cost control strategies; The intelligent decision-making unit combines the current real-time status of the project and the reserve data of the knowledge base unit to generate a customized cost control strategy combination through a preset algorithm.

[0016] Detailed steps for generating an adaptive adjustment strategy: The system's built-in massive knowledge base storage structure utilizes a distributed database combined with a knowledge graph to facilitate rapid retrieval and correlation analysis. Details of past project responses cover a full range of scenarios, from minor engineering changes to major force majeure events. For example, during a high-rise building construction season encountering strong winds, successful measures such as pre-emptive crane reinforcement and adjustments to the exterior wall construction sequence were implemented. Data evaluating the effectiveness of various cost control strategies in different scenarios is recorded through quantitative indicators such as cost reduction, duration of construction delays, and changes in project quality scores, providing a reliable reference for the development of new strategies.

[0017] In the event of a sudden increase in raw material prices, when generating a strategy based on the cost optimization objective function, the system first conducts a comprehensive screening of alternative materials, considering comprehensive factors such as whether the performance of the alternative materials meets the project requirements, supply stability, procurement costs, and transportation costs. In terms of adjusting the construction sequence, the process cost sensitivity function is used to accurately identify processes that are less affected by price fluctuations and prioritize construction. For example, in prefabricated building construction, if the price of prefabricated components rises, but the on-site foundation construction process is less affected and has low cost sensitivity, resources can be concentrated on basic construction first, while existing inventory materials can be rationally allocated. The inventory management system can be used to query the inventory, shelf life, and other information of materials in different warehouses in real time, delaying high-priced purchases and achieving dynamic cost optimization. At the same time, it provides managers with detailed effect estimates after the implementation of each strategy, such as the estimated amount of cost reduction and the estimated number of days of change in construction period, to assist in efficient decision-making.

[0018] Detailed explanation of the data acquisition module in this embodiment: The tower crane load sensor in the sensor unit uses a strain gauge sensor with high precision and high stability. It can adapt to the vibration and impact environment of high-intensity lifting operations of tower cranes and reliably measure material weight. Environmental sensors include temperature and humidity sensors, air pressure sensors, wind speed and direction sensors and other types. They are distributed in key areas of the construction site, such as concrete pouring areas and high-altitude work surfaces, to collect meteorological data in all directions. Equipment operation status sensors monitor the equipment's current, voltage, vibration frequency and other parameters to grasp the operating status of key equipment such as cranes, mixers, and welding machines in real time, provide early warning of equipment failure risks, and reduce the cost increase caused by sudden equipment failures.

[0019] When connecting to supplier systems, the data interface unit adheres to mainstream security authorization protocols such as OAuth 2.0 to ensure confidentiality and integrity during data transmission. Furthermore, data caching and resumable transfer capabilities prevent data loss due to network fluctuations, ensuring continuous and stable access to key data such as raw material prices.

[0020] The manual data entry unit's mobile app interface features a simple and intuitive form layout, complemented by auxiliary functions such as voice input and image upload. When construction workers encounter special circumstances, such as the discovery of cultural relics requiring work stoppage for protection or the rerouting of underground pipelines, they can quickly complete information entry by describing the incident in voice and uploading on-site photos, allowing the system to keep abreast of on-site dynamics.

[0021] Detailed explanation of the data analysis and modeling module in this embodiment: The data cleaning algorithm in the data processing unit utilizes a hybrid model based on statistical rules and machine learning. Statistical rules are first used to identify obvious outliers, such as data that exceeds the normal material usage range by several times. Unsupervised learning algorithms are then used to perform cluster analysis on suspected but unclear data to determine whether it represents a new construction model or reasonable data under special working conditions. Outliers are accurately removed, missing values ​​are filled, and heterogeneous data is standardized into a standard format, providing high-quality data for modeling.

[0022] The machine learning algorithm library integrated into the model building and optimization unit covers a variety of classic algorithms, including decision trees, neural networks, and support vector machines, and automatically selects adaptive algorithms based on project characteristics. For cost prediction scenarios with strong linear relationships, such as the relationship between material cost and schedule for simple building structures, a linear regression algorithm may be preferred to build the initial model. For complex nonlinear relationships, such as large-scale comprehensive buildings involving multiple types of work and multiple materials that interact to affect costs, a neural network algorithm is used. The backpropagation algorithm continuously fine-tunes the model parameters, continuously improving the model's adaptability and prediction accuracy for complex engineering scenarios based on real-time feedback data. The model building and optimization process strictly adheres to the aforementioned cost prediction model formula and optimization methods.

[0023] Detailed explanation of the monitoring and early warning module: The real-time comparison unit acquires data as frequently as once per second, ensuring timely capture of even the slightest cost fluctuations. A combination of interpolation and trend analysis is employed to compare actual and forecasted cost data. This approach not only focuses on current cost differences but also analyzes cost trends through fitting past cost data. This allows for early detection of potential cost overruns and pinpoints the source of cost fluctuations, such as inefficient construction teams or delayed delivery from a particular material supplier. This provides a precise basis for early warning decisions.

[0024] When sending alerts, the early warning push unit sets different push levels based on the responsibilities and permissions of management personnel. For project managers, detailed cost deviation analysis reports, potential risk points, and response recommendations are sent; for frontline construction supervisors, concise and clear early warning information on key cost indicators, such as reminders of material cost thresholds exceeding thresholds and labor hours exceeding standards, is sent. Alerts are delivered promptly and without omission through multiple channels, including SMS, system pop-ups, and in-site messages. The information formats of different channels are adapted to the characteristics of the receiving terminal, such as concise text in SMS and intuitive charts in system pop-ups.

[0025] Detailed explanation of the strategy generation module: The knowledge base stores a vast amount of historical project cost control cases, categorized and indexed by project type, region, scale, and timeframe, enabling quick and easy access to similar cases. For example, the cost control experience of similar high-rise buildings in East China over the past five years during typhoon season can be searched, providing targeted reference for current projects. Detailed implementation steps for various cost control strategies are documented, including information on everything from policy activation conditions, resource allocation processes, construction process adjustment details, and acceptance criteria. Results feedback covers quantitative indicators across multiple dimensions, including cost, construction period, quality, and safety, laying a solid foundation for developing new strategies.

[0026] The intelligent decision-making unit combines the current project's real-time status with the knowledge base, employing an intelligent reasoning algorithm that combines case-based reasoning (CBR) and rule-based reasoning (RBR). CBR first retrieves similar cases from the knowledge base and extracts preliminary strategies. RBR then applies rule matching to the specific conditions of the current project, such as contractual constraints, available on-site resources, and surrounding environmental restrictions, to optimize and adjust strategies. Within minutes, the unit rapidly generates multiple customized cost control strategy combinations, accompanied by detailed implementation effect estimates, to assist management in making optimal decisions based on on-site conditions. Strategies are generated based on the aforementioned cost optimization objective function and process cost sensitivity function.

[0027] Compared with the prior art, the present invention has the following beneficial effects: (1) This invention uses big data analysis technology and machine learning algorithms to build a cost prediction model, taking into account factors such as the labor efficiency variation coefficient at different construction stages, the downtime caused by abnormal weather and the additional cost coefficient caused by it. It can accurately reflect the cost change trend and overcome the problem of inaccurate cost prediction caused by traditional methods due to information limitations and lack of dynamic analysis; (2) The present invention monitors the project cost in real time based on the constructed model. Combining the project characteristics with the cost budget planning, the present invention sets independent thresholds for the material cost and labor cost of key construction nodes. The thresholds can be dynamically adjusted according to the actual situation of the project. When the cost forecast value fluctuates abnormally, such as the material cost of a key construction stage rises by more than 10% in a short period of time or the labor cost deviates from the normal range by a certain proportion, it can immediately trigger a precise early warning, timely discover the cost risk, and change the passive situation of post-adjustment of the traditional method. (3) The system of the present invention has a built-in massive knowledge base and intelligent algorithms. At the moment of early warning triggering, it can quickly generate multiple sets of targeted strategies based on the experience of coping with similar engineering cases in the past and the evaluation data of the implementation of various cost control strategies. For example, when the price of raw materials rises sharply, it can recommend the urgent expansion and screening of high-quality alternative material suppliers, cleverly adjust the construction sequence to give priority to the processes that are less affected by price fluctuations, and reasonably allocate existing inventory materials according to the process cost sensitivity function to delay high-priced purchases, etc. It also provides an estimate of the effect of each strategy after implementation, assists managers in making efficient decisions, and effectively solves the dilemma of traditional methods lacking coping strategies when cost problems arise, improves the level of cost management of construction projects, avoids cost overruns and waste of resources, and ensures the economic benefits of the project.

Claims

1. An adaptive construction project cost control method, characterized in that: The method comprises the following steps: S1: Monitor the weight of lifted materials , combined with the number of lifts , calculate the real-time weight of the material by the following formula: ； in, The weight of the material lifted each time; Collect weather conditions data and real-time price data of raw materials , set up a manual input port to collect time-consuming data of special construction processes ; S2: Introduce big data analysis to deeply process the collected data and build a system based on the project progress. The horizontal axis shows cost changes. The vertical axis is the cost prediction model, which takes into account the change coefficient of labor efficiency in different construction stages. , the length of downtime caused by abnormal weather and the additional cost coefficient it causes , set up a cost prediction model and continuously optimize the model through machine learning algorithms so that it can dynamically calibrate the prediction results based on real-time inflow data; S3: Based on the cost prediction model set in step S2, the project cost is monitored in real time. Combining the project characteristics with the cost budget planning, independent thresholds are set for the material cost and labor cost of the preset key construction nodes. , labor costs ; When the material cost forecast value of a preset key construction stage is If the upward trend meets the preset formula, or the labor cost deviates from the normal pickup rate by a preset ratio, an early warning will be triggered: S4: At the moment the warning is triggered, the system selects higher-quality alternative material suppliers based on the built-in knowledge base and algorithm, adjusts the construction sequence to prioritize processes that are relatively less affected by price fluctuations, and then allocates existing inventory materials based on the process cost sensitivity function, providing an estimate of the effects of each strategy after implementation.

2. The adaptive construction project cost control method according to claim 1, characterized in that: In step S2, the cost prediction model is: ； in, Indicates different construction stages.

3. The adaptive construction project cost control method according to claim 1, characterized in that: In step S3, the preset formula is: ； in, for The predicted value of material cost at the moment, for The forecast value of material cost at the moment.

4. The adaptive construction project cost control method according to claim 1, characterized in that: In step S4, the sensitivity function is: ； in, For the A process, is the total cost.

5. An adaptive construction project cost control system, characterized in that: The system includes a data acquisition module, a data analysis and modeling module, a monitoring and early warning module, and a strategy generation module; The data acquisition module includes a sensor unit, a data interface unit and a manual input unit; The sensor unit includes a tower crane load sensor, an environmental sensor, and an equipment operation status sensor. The sensor unit is used to convert physical quantities at the construction site into electrical signals or digital signals, and capture preset key data parameters in real time; The data interface unit is used to establish a dedicated data interface with various supplier systems including those for building materials and equipment rental; The manual input unit uses a human-computer interaction interface to allow frontline construction workers to input working condition information into the system via a mobile terminal or a fixed terminal; The data analysis and modeling module includes a data processing unit and a model building and optimization unit; The data processing unit has built-in data cleaning, conversion and aggregation algorithms to pre-process the heterogeneous data in the system, remove outliers, fill in missing values, and regularize the processed data into a unified standard format; The model building optimization unit integrates a machine learning algorithm library, automatically screens suitable algorithms based on engineering characteristics and builds an initial cost prediction model. At the same time, the model building optimization unit continuously adjusts model parameters based on real-time feedback data; The monitoring and early warning module includes a real-time comparison unit and an early warning push unit; The real-time comparison unit is configured to continuously obtain actual cost data and predicted cost data output by the data analysis and modeling module, and compare the differences to locate the source of cost fluctuations; The warning push unit pushes alarm information to project managers at all levels according to preset warning rules when the cost deviation reaches a threshold; The strategy generation module includes a knowledge base unit and an intelligent decision-making unit; The knowledge base unit is configured to store historical project cost control cases, including information such as project type, project scale, success records and failure records, as well as implementation steps, resource inputs and effectiveness feedback data of historical cost control strategies; The intelligent decision-making unit combines the current real-time status of the project and the reserve data of the knowledge base unit to generate a customized cost control strategy combination through a preset algorithm.

6. The adaptive construction project cost control system according to claim 5, characterized in that: In the model construction optimization unit, the construction is based on the engineering progress The horizontal axis shows cost changes. The vertical axis is the cost prediction model, which takes into account the change coefficient of labor efficiency in different construction stages. , the length of downtime caused by abnormal weather and the additional cost coefficient it causes , the cost prediction model is: ； in, Indicates different construction stages.

7. The adaptive construction project cost control system according to claim 5, characterized in that: In the intelligent decision-making unit, the strategy generation basis is the cost optimization objective function and the process cost sensitivity function, and the sensitivity function is: ； in, For the A process, is the total cost.

Images