Intelligent construction site data processing method based on BIM

Abstract

The invention discloses an intelligent construction site data processing method based on a BIM, and relates to the BIM, and the method comprises the steps: obtaining the basic data of a construction site through a BIM model, and determining a to-be-optimized construction resource; the construction site data collected in real time is compared with basic data obtained in the BIM model, and a key deviation value is calculated; establishing a construction resource optimization model according to the key deviation value; solving the construction resource optimization model by using an improved genetic algorithm to obtain an optimal construction resource configuration scheme; the improved genetic algorithm comprises the following steps: generating an initial solution set according to a BIM model; dividing the population into a first region and a second region, performing global search in the first region by adopting a high variation rate, and accelerating local optimization in the second region by adopting an elitism strategy; setting a coevolution mechanism between the first region and the second region; for the low resource utilization rate of the intelligent construction site, data are integrated through a BIM model, process logic and storage space constraints are considered, and the optimization efficiency is improved by adopting an improved double-region co-evolution genetic algorithm.

Description

Technical Field

[0001] This application relates to the field of BIM, and particularly to a BIM-based intelligent construction site data processing method. Background Art

[0002] Hospital projects involve the simultaneous scheduling of ordinary construction resources (such as concrete, steel bars) and special medical resources (such as medical equipment, clean systems). These different types of resources have different optimization objectives and constraints. Ordinary construction resources mainly consider cost and schedule, while medical equipment, especially large-scale precision equipment (such as magnetic resonance imaging machines, CT scanners, etc.), has extremely high requirements for installation environment, timing arrangement, and logistics support.

[0003] On the one hand, traditional methods often optimize ordinary construction resources and medical equipment resources separately, resulting in poor overall coordination and often causing process conflicts and resource waste. For example, after the completion of the building main body, it is found that the reserved holes do not match the large medical equipment, or when the precision equipment arrives, it is found that the construction environment does not meet the standards. Existing methods have poor effects in dealing with complex constraints such as limited material storage space and multi-process cross-operation, and especially lack effective support for the construction resource allocation in special areas such as hospital clean areas and radiation protection areas.

[0004] On the other hand, as a heuristic optimization method, the genetic algorithm has been widely used in the optimization of construction resources. However, the traditional genetic algorithm still has obvious deficiencies in dealing with complex resource scheduling problems such as hospital projects: the traditional genetic algorithm often adopts a single population evolution strategy, which cannot take into account both global search and local optimization at the same time. When dealing with complex optimization problems with multiple objectives and multiple constraints, it is easy to fall into local optimal solutions or have a too slow convergence speed. In most applications, the genetic algorithm generates the initial solution randomly, without making full use of the engineering information in the BIM model, resulting in a low quality of the initial population and limited optimization efficiency.

[0005] Therefore, based on an effective intelligent construction site data processing method, the construction efficiency and resource utilization rate of hospital projects are improved. Summary of the Invention

[0006] Aiming at the low resource utilization rate of the intelligent construction site, this application provides a BIM-based intelligent construction site data processing method, which integrates data through the BIM model, considers process logic and storage space constraints, and adopts an improved two-region co-evolution genetic algorithm to improve the optimization efficiency.

[0007] This application provides a BIM-based intelligent construction site data processing method, including: obtaining the basic data of the construction site using a BIM model and determining the construction resources to be optimized; the basic data includes design parameters, design construction periods, and process dependencies; the construction resources to be optimized include material supply; among them, the design parameters refer to the building and engineering technical indicators extracted from the BIM model and are the basic input data for optimizing construction resources. It can include: geometric parameters: spatial information such as the size, shape, area, and volume of the building; material parameters: information such as the material type, specification, and performance requirements of each component; equipment parameters: technical parameters of medical equipment unique to the hospital, such as size, weight, installation requirements, and environmental condition requirements; the process dependency refers to the sequence and logical connection existing between each process during the construction process and is the basic constraint that must be followed for optimizing construction resources. For example, the competition relationship of multiple processes for the same resource, such as multiple areas simultaneously requiring the use of tower cranes; the sequence of multiple processes implemented in the same space, especially the spatial coordination of the installation of dense pipelines in the hospital; special process dependencies such as clean areas and radiation protection areas unique to the hospital, etc.

[0008] Compare the real-time collected construction site data with the basic data obtained from the BIM model, calculate the key deviation values, and the key deviation values include construction progress deviation and resource utilization rate deviation; according to the key deviation values, establish a construction resource optimization model, and the construction resource optimization model aims to minimize the deviation value and sets construction constraint conditions according to the basic data; use an improved genetic algorithm to solve the construction resource optimization model to obtain the optimal construction resource allocation plan.

[0009] Among them, the improved genetic algorithm includes: generating an initial solution set according to the BIM model, and the initial solution set satisfies the key construction constraints, and the key construction constraints include process logic constraints; dividing the population into a first area and a second area, using a high mutation rate for global search in the first area and using an elitist retention strategy to accelerate local optimization in the second area; specifically, according to the distribution of fitness values, the initial population can be divided into two parts: individuals with lower fitness but higher diversity are assigned to the first area for global search, while individuals with higher fitness are assigned to the second area for local optimization. It can also be in proportion, usually the first area accounts for 40% - 60% of the population, and the second area accounts for the remaining part. Set the co-evolution mechanism between the first area and the second area.

[0010] Further, compare the real-time construction site data collected with the basic data obtained from the BIM model, and calculate the key deviation values, including: collect real-time construction site data using the Internet of Things devices deployed at the construction site, and the real-time construction site data includes construction progress data and resource consumption data; structure the real-time construction site data according to the component coding system of the BIM model to associate the spatial location relationship between the real-time construction site data and the basic data of the BIM model; calculate the construction progress deviation value by comparing the actual construction progress with the designed construction period of the BIM model; calculate the resource utilization rate deviation value by comparing the material consumption with the designed parameters of the BIM model.

[0011] Further, establish a construction resource optimization model based on the key deviation values, including: construct a multi-objective optimization function according to the construction progress deviation value and the resource utilization rate deviation value; extract construction constraint conditions according to the basic data of the BIM model, and the construction constraint conditions include: material storage space constraint; among them, the material storage space constraint refers to the limitation condition of the construction site material storage capacity that must be considered during the resource optimization of the smart construction site. Its mathematical expression is: , this constraint ensures that the total space occupied by all materials at any time point t does not exceed the total available storage space at the construction site at this time point.

[0012] Further, the multi-objective optimization function is: , where represents the construction progress deviation value of the i-th process, represents the resource utilization rate deviation value of the j-th type of material, and are weight coefficients and satisfy ; the material storage space constraint is: , where represents the storage quantity of the k-th type of material at time t, represents the storage space occupied by a unit of the k-th type of material, represents the total available storage space at the construction site at time t.

[0013] Specifically, by incorporating the progress deviation ( ) and the resource utilization rate deviation ( ) into the same function, a unified optimization framework for ordinary building resources and special medical resources is realized, and the process conflict problem caused by separate optimization of traditional methods is solved. In addition, by introducing the time dimension t, the static space constraint is transformed into a dynamic constraint, enabling the system to handle the characteristics of the construction site storage capacity changing with the construction stage, and avoiding the resource scheduling conflict caused by ignoring the time sequence change in the traditional model. Finally, through The parameter differentiates the unit space occupancy of different materials, accurately expressing the huge difference in space requirements between ordinary building materials and large medical equipment (such as magnetic resonance imaging (MRI) machines and CT scanners), laying the foundation for subsequent adjustment of the equipment sensitivity level (SE).

[0014] Furthermore, an improved genetic algorithm is used to solve the construction resource optimization model to obtain the optimal construction resource allocation plan, including: encoding the construction resource allocation plan as a chromosome, where the chromosome contains the material supply time and quantity; constructing a fitness function based on the multi-objective optimization function; generating an initial solution set according to the BIM model as the initial population; dividing the population into a first region and a second region: in the first region, a mutation operation with a mutation rate greater than the threshold is used for global search, and the mutation operation is to randomly adjust the material supply time and quantity to explore a larger solution space; in the second region, a mutation rate less than the threshold and the elite retention strategy are used for local optimization, and the top P% of individuals with the highest fitness in each generation are directly selected to enter the next generation; a co-evolution mechanism between the first region and the second region is set, and the co-evolution mechanism balances the exploration and exploitation of the algorithm by periodically exchanging individuals between regions and adjusting the fitness of migrating individuals; when the iteration termination condition is met, the optimal individual is output as the optimal construction resource allocation plan.

[0015] Furthermore, the fitness function is: , where SE is the equipment sensitivity level, the SE value of large precision equipment such as MRI machines and CT scanners is greater than 1, and the SE value of ordinary medical equipment is equal to 1, is the weight coefficient; F is the value of the multi-objective optimization function.

[0016] Furthermore, generating the initial solution set according to the BIM model includes: extracting the process dependency relationship and material requirement information from the BIM model to construct a process network model in the form of a directed acyclic graph; calculating the material types and quantities required for each process according to the construction resources to be optimized to form a material requirement matrix M; generating an initial material supply plan through heuristic rules according to the process network model and the material requirement matrix M, and the initial supply plan contains the supply time points for each batch; correcting the initial supply plan by moving the supply time points forward and backward to meet the process dependency relationship; checking whether the corrected supply plan meets the material storage space constraint , if not, then correct the supply plan again until all time points meet the storage space constraint; encoding the corrected supply plan as a chromosome as the initial population.

[0017] Specifically, on the one hand, traditional genetic algorithms generally generate the initial population in a completely random manner without considering the constraint conditions and professional knowledge in the problem domain, resulting in a low-quality population due to random initialization. In this application, by extracting the process dependency relationship and material requirement information from the BIM model, a process network model in the form of a directed acyclic graph (DAG) is constructed. The individuals in the initial population already satisfy the key construction constraints, avoiding the generation and evaluation of a large number of invalid solutions in traditional algorithms, and reducing the search space from the theoretically possible set to the practically feasible set in engineering. Through the topological sorting property of the directed acyclic graph, it is ensured that the initial supply plan is reasonably generated on the premise of satisfying the process logic, avoiding a large number of logical contradictions that may be introduced by random search.

[0018] On the other hand, when dealing with multi-objective and multi-constraint problems, traditional algorithms generally adopt a single-population evolution strategy and a simple penalty function mechanism. Therefore, they are prone to falling into local optimal solutions, resulting in slow convergence speed and low optimization efficiency in complex scenarios such as hospital projects. In this application, through structured diversity design and progressive constraint correction methods, not only the quality of the initial solution is improved, but also high-quality genetic materials are provided for the subsequent two-region co-evolution. At the same time, this method establishes a bridge between engineering domain knowledge and computational optimization by converting the engineering semantics in the BIM model into chromosome encoding of the genetic algorithm, enabling the algorithm to directly use the professional knowledge of hospital projects to guide the optimization process, fundamentally improving the blindness and inefficiency of traditional genetic algorithms in dealing with complex engineering optimization problems.

[0019] Furthermore, in the first region, a mutation operation with a mutation rate greater than the threshold is used for global search. The mutation operation randomly adjusts the material supply time and quantity to explore a larger solution space, including: setting the mutation rate threshold , and randomly generating the actual mutation rate v in each mutation operation, such that v ∈ [0.15, 0.3]; for the individuals in the first region of the initial population, randomly select the gene position of the material supply time on the chromosome for mutation, and the mutation probability is v; for the material supply time represented by the gene position, perform a random offset within the range allowed by the obtained process dependency relationship, and the offset amount D ∈ [-5, 5] time units; perform a process logic constraint check on the mutated individual. If the process logic constraint is violated, find the earliest feasible time point after the completion of the previous process in the process network model as the effective time point, and adjust the supply time to the nearest effective time point; perform a material storage space constraint check on the mutated individual. If there is a violation, reduce the material supply quantity at the violation time point and correspondingly increase the supply quantity at the adjacent time point; use the fitness function Evaluate the fitness value of the individuals after mutation, and retain the individuals with improved fitness after mutation. For the individuals with decreased fitness, retain them with a probability of P1% to maintain the population diversity.

[0020] Specifically, the traditional genetic algorithm faces the "exploration - exploitation" dilemma, that is, the algorithm either focuses on exploring new solution spaces (possibly missing local optimization opportunities) or focuses on exploiting known solutions (easily falling into local optima). This solution realizes the parallel promotion of exploration and exploitation at the algorithm level through the regional differentiation strategy, significantly improving the global optimization ability of the algorithm. In addition, the random offset within the range of [-5, 5] is not blind, but is executed within the range allowed by the process dependency relationship, and is constrained and repaired through the process network model, achieving efficient exploration while ensuring the feasibility of the solution and avoiding a large number of ineffective searches. Finally, different correction strategies are adopted for individuals that violate the process logic constraints and material storage space constraints, reflecting the refined processing of different types of constraints, which is particularly important for complex scenarios where ordinary building resources and special medical equipment coexist in hospital engineering.

[0021] Furthermore, in the second region, a mutation rate less than the threshold and the elite retention strategy are used for local optimization. Select the top P% of the individuals with the highest fitness in each generation to directly enter the next generation, including: setting the mutation rate threshold , and randomly generating the actual mutation rate during each mutation operation, so that v' ∈ [0.01, 0.05]; for the individuals in the second region of the initial population, sort them in descending order of fitness value, and select the top P2% of the individuals with the highest fitness as elite individuals to directly enter the next generation; for the remaining individuals, select the gene locus of the material supply time on the chromosome for mutation, and the mutation probability is ; for the material supply time represented by the selected gene locus, perform an offset within the range allowed by the process dependency relationship, and the offset amount D' ∈ [-2, 2] time units; perform a process logic constraint check on the mutated individuals. If there is a violation, find the earliest feasible time point after the completion of the previous process in the process network model as the valid time point, and adjust the supply time to the nearest valid time point; perform a material storage space constraint check on the mutated individuals. If there is a violation, adjust the supply quantity at multiple time points; use the fitness function to evaluate the fitness value of the mutated individuals, and only retain the individuals with improved fitness after mutation, and eliminate the individuals with decreased fitness to enhance the local optimization efficiency.

[0022] Specifically, set the mutation rate threshold and randomly generate The actual mutation rate is significantly lower than that of the first region [0.15, 0.3]. This fine mutation control enables the algorithm to deeply explore the potential solution space region, avoiding the dilemma in traditional genetic algorithms where a single mutation rate leads to either excessive perturbation that destroys high-quality solutions or insufficient mutation that makes it difficult to improve. In addition, the mechanism of directly selecting the top P2% individuals with the highest fitness to enter the next generation ensures that the algorithm will not lose the high-quality solutions that have been discovered, solving the problem of easily losing the historical optimal solutions in traditional genetic algorithms and providing a stable optimization path for complex constraint problems.

[0023] The first region adopts a high mutation rate (v ∈ [0.15, 0.3]) and a lenient selection strategy, focusing on global exploration; the second region adopts a low mutation rate and a strict selection strategy, focusing on local development; the two regions achieve information sharing and improved optimization efficiency by periodically exchanging high-quality individuals and diverse individuals; this two-region co-evolution strategy solves the core contradiction of traditional single-population genetic algorithms - the inability to balance the breadth of the search space and the convergence speed simultaneously, and is particularly suitable for complex optimization scenarios such as hospital engineering with heterogeneous resource allocation requirements. By ensuring that the algorithm can deeply explore the high-quality solutions that have been discovered while maintaining sufficient exploration ability, a high-quality optimization plan that can be implemented in engineering practice is finally obtained.

[0024] Furthermore, a co-evolution mechanism between the first region and the second region is set up. The co-evolution mechanism balances the exploration and development of the algorithm by periodically exchanging individuals between regions and adjusting the fitness of the migrating individuals, including: setting the interval exchange period T, and performing interval exchange once every T generations of iteration, where T is P3% of the number of iterations; calculating the gradient diversity index MD of the individuals in the first region, and selecting the top P4% individuals with the highest MD values as diversity representatives; selecting the top P5% individuals with the highest fitness from the elite individuals in the second region as high-quality solution representatives; migrating the diversity representatives selected from the first region to the second region, and migrating the high-quality solution representative individuals selected from the second region to the first region; adjusting the adaptability of the individuals migrated from the first region to the second region by reducing the random perturbation range of the corresponding individual's material supply time to [-1, 1]; enhancing the exploration ability of the individuals migrated from the second region to the first region by increasing the mutation probability of the corresponding individual to v + 0.1 and expanding the random perturbation range of the material supply time to [-7, 7]; re-evaluating the populations of the two regions after migration, updating the fitness ranking, and preparing for the next generation of iteration.

[0025] Specifically, this application simulates the gene exchange phenomenon among different ecological niche groups in biological evolution, and upgrades the single-population genetic algorithm to a multi-population co-evolution system. This bionic idea enables the algorithm to maintain both "exploration" and "exploitation" evolutionary pressures simultaneously, solving the convergence dilemma caused by the single evolutionary strategy in traditional genetic algorithms. In addition, for the individuals migrating from the first region to the second region, the perturbation range is reduced to [-1, 1] to promote their transformation from "explorers" to "optimizers"; for the individuals migrating from the second region to the first region, the mutation probability is increased and the perturbation range is expanded to [-7, 7] to stimulate their exploration potential.

[0026] Compared with the prior art, the advantages of this application are as follows:

[0027] Hospital engineering involves the simultaneous scheduling of ordinary construction resources (concrete, steel bars) and special resources (medical equipment, clean systems). Different resource types have different optimization objectives and constraints. Traditional methods often optimize separately, resulting in overall disharmony. In addition, traditional genetic algorithms often adopt a single-population evolutionary strategy, which cannot simultaneously balance global search and local optimization, is prone to falling into local optimal solutions or having a slow convergence speed, and lacks consideration of the sensitivity of special equipment, resulting in unreasonable resource allocation and poor optimization effects.

[0028] On the one hand, this application integrates multi-source data through a BIM model and constructs a multi-objective optimization function considering material storage space constraints, realizing the unified collaborative optimization of ordinary building resources and special medical resources. On the other hand, through an improved two-region co-evolution genetic algorithm, the population is divided into a global search area with a high mutation rate and a local optimization area with elite retention, and an interval exchange mechanism is set up to achieve a balance between the search space and the optimization efficiency. At the same time, a device sensitivity parameter is introduced to adjust the fitness function, enabling the algorithm to make more reasonable optimization decisions for special resources such as medical precision equipment, thus significantly improving the overall resource utilization rate and construction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] This application will be further described in the form of exemplary embodiments, which will be described in detail through the drawings. These embodiments are not restrictive. In these embodiments, the same numbers represent the same structures, where:

[0030] Figure 1 is an exemplary flowchart of a BIM-based intelligent construction site data processing method shown in some embodiments of this application;

[0031] Figure 2 is an exemplary flowchart of generating an initial population shown in some embodiments of this application;

[0032] Figure 3 is an exemplary flowchart of the iteration of individuals in the first region shown in some embodiments of this application;

[0033] Figure 4 It is an exemplary flowchart of the second - area individuals shown in some embodiments of the present application. Detailed implementation manners

[0034] The methods and systems provided by the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0035] As Figure 1 shown, use the BIM model to obtain the basic data of the construction site and determine the construction resources to be optimized; the basic data includes design parameters, design duration, and process - dependency relationships; the construction resources to be optimized include material supply; compare the real - time collected construction - site data with the basic data obtained from the BIM model, calculate the key deviation values, and the key deviation values include construction - progress deviation and resource - utilization deviation; according to the key deviation values, establish a construction - resource optimization model, and the construction - resource optimization model aims to minimize the deviation values and sets construction constraint conditions according to the basic data; use an improved genetic algorithm to solve the construction - resource optimization model to obtain the optimal construction - resource allocation plan; among them, the improved genetic algorithm includes: generating an initial solution set according to the BIM model, and the initial solution set satisfies the key construction constraints, and the key construction constraints include process - logic constraints; dividing the population into a first area and a second area, performing global search with a high mutation rate in the first area and accelerating local optimization with an elitist - retention strategy in the second area; setting a co - evolution mechanism between the first area and the second area.

[0036] Obtain BIM model files in Revit, ArchiCAD, or Navisworks format from hospital engineering design units; establish a multi - level coding system including functional areas, specialties, and component types according to the "Classification and Coding Standard for Building Information Models"; perform special coding for special medical areas such as operating rooms, radiology departments, and nuclear medicine departments; establish a coding mapping relationship between medical equipment and BIM - model components. Extract geometric parameters, especially extract the actual dimensions and installation - space requirements of large medical equipment (such as magnetic resonance imaging machines, CT scanners), extract the positions and dimension parameters of various reserved holes, etc.

[0037] Extract design - duration data from BIM - model appendices or project - management systems, identify and extract key milestone nodes and time control points, decompose the total duration into sub - durations by area, specialty, and system, and extract the duration - constraint requirements for special processes and medical - equipment installation, etc.

[0038] Based on the spatial positions and physical connections of components, automatically deduce the preliminary process - dependency relationships, and supplement the logical dependencies that cannot be directly deduced from the physical relationships by engineering experts. Especially establish the process - dependency relationships between medical equipment and building projects, and generate a complete process - dependency relationship matrix, etc.

[0039] Deploy a multi-level Internet of Things (IoT) perception system at the construction site to collect construction progress data and resource consumption data: Use a combination of RFID tags and QR code recognition to mark key components and process nodes. Workers scan the completion status through handheld terminals, and at the same time, fixed cameras are deployed in important areas for visual recognition. The system automatically summarizes the progress data every 4 hours, and real-time monitoring is achieved for key processes such as the installation of medical equipment.

[0040] Convert the collected heterogeneous data into structured data that can be docked with the BIM model: Based on the ISO16739 standard, construct a hierarchical coding structure, including: project ID - professional code - floor number - component type - component number. For example, "MH001-MEP-03-EQP-0021" represents the medical equipment on the third floor.

[0041] The progress deviation calculation adopts a multi-index comprehensive evaluation method: The calculation of the process completion rate (CR): = The number of components of process i actually completed / The number of components of process i that should be completed; The calculation of the time deviation (TD): = The actual completion time - The planned completion time; The comprehensive calculation formula of the progress deviation value (PD): , where is the maximum acceptable time deviation, and are the weight coefficients.

[0042] The resource utilization rate deviation calculation adopts a hierarchical classification comparison method: The calculation of the material consumption efficiency (MCE): = The actual completed workload / The actual consumed material quantity j ÷ The standard workload / The standard material consumption quantity j. The calculation of the resource utilization rate (RU): = The actual material usage quantity j / The designed material demand quantity j × 100%. The calculation of the resource utilization rate deviation value (RD): , where is the material importance coefficient, and the β value of medical special equipment is higher than that of ordinary building materials.

[0043] According to the key deviation values, establish a construction resource optimization model, including: According to the construction progress deviation value and the resource utilization rate deviation value, construct a multi-objective optimization function; , where represents the construction progress deviation value of the i-th process, represents the resource utilization rate deviation value of the j-th type of material, and are the weight coefficients, and satisfy ;

[0044] Extract construction constraint conditions based on the basic data of the BIM model. The construction constraint conditions include: material storage space constraint; the material storage space constraint is: , where represents the storage quantity of the k-th type of material at time t, represents the storage space occupied by a unit of the k-th type of material, represents the total available storage space at the construction site at time t;

[0045] As Figure 2 shown, this solution uses a hybrid coding technique to transform the construction resource allocation plan into a computable chromosome structure: each chromosome consists of two parallel sequences: the material supply time sequence and the corresponding quantity sequence. The time sequence uses integer coding, representing the time unit relative to the start of the project; the quantity sequence uses floating-point coding, representing the quantity of materials supplied each time. Considering the characteristics of multiple material types in hospital projects, a hierarchical chromosome structure is adopted, and different types of materials occupy different sections in the chromosome. An independent coding section is specially set for precision medical equipment.

[0046] This solution designs an innovative fitness function considering the sensitivity of medical equipment: , where F is the multi-objective optimization function value, covering schedule and resource deviation, SE is the equipment sensitivity level, introducing the characteristics of medical equipment, is the adjustment coefficient, controlling the influence intensity of SE on the fitness. Considering the particularity of hospital projects, the equipment sensitivity level SE parameter is introduced: for large-scale precision equipment such as magnetic resonance imaging (MRI) and CT scanners: SE > 1 (usually set to 1.5 - 2.0); for ordinary medical equipment: SE = 1; for conventional building materials: the SE parameter is not explicitly introduced.

[0047] Generate a high-quality initial solution set based on the BIM model. The data parser based on the IFC standard supports the IFC2x3 and IFC4 standards, and directly extracts the precedence relationship of processes from the 4D - BIM time information, in graph structure: , representing the process set; , process must be completed before process . Each process node contains key attributes, process ID and name, planned start time and duration, resource type requirement list, process priority and sensitivity mark.

[0048] Extract material parameters (specifications, quantities, performance) from the component attributes of the BIM model, integrate the resource consumption quotas in the project management system, and construct a material demand matrix, , where represents the demand quantity of the i-th process for the j-th type of material, n is the total number of processes, and k is the total number of material types. Perform time expansion on the demand matrix, , representing the demand of the $i$-th process for the $j$-th type of material at time $t$.

[0049] Set a multi-objective heuristic rule set. Critical path priority rule: prioritize meeting the material requirements of processes on the critical path; balanced supply rule: avoid spikes in the supply curve and smooth resource allocation; latest supply rule: supply materials as close as possible to the usage time to reduce storage pressure; batch optimization rule: determine the economically reasonable supply batch according to material characteristics; equipment priority guarantee rule: prioritize determining the supply plan for precision medical equipment.

[0050] Calculation of process priority: ; $CP$ is the critical path index (0 or 1); $TO$ is the process topological order; $SE$ is the equipment sensitivity level; is the weight coefficient. Generate the initial supply plan, the supply time point $t$ = process start time - lead coefficient $\beta$, where $\beta = f$ (material type, storage conditions, usage urgency). Batch supply quantity , where $\gamma$ is the material loss coefficient, determined according to material type and project experience.

[0051] Detection of violation of process dependency constraints. For any process and its preceding process : If $t$ (supply, $j$) < $t$ (complete, $i$), then the process constraint is violated. If the supply time point violates the process dependency: $t$ (supply, $j$) = max ($t$ (supply, $j$), $t$ (complete, $i$) + safety interval).

[0052] Inspection of storage space constraints, $C_t$ = base capacity ± phased capacity adjustment. Calculate the available storage space at each time point based on the 4D - BIM site layout information. The occupied space at time $t$ = , = ∑ (supply quantity) - ∑ (consumption quantity) for all $k$ types of materials up to time $t$. For those that violate the storage space constraint inspection, large batches can be split into multiple small batches to disperse the supply pressure.

[0053] As Figure 3 shown, the first area adopts a high-strength mutation strategy, the purpose of which is to fully explore the decision space, discover potential high-quality solution regions, and prevent the algorithm from falling into local optima. Set the benchmark mutation rate threshold , and use the random floating technique to generate the actual mutation rate $v\in[0.15, 0.3]$. When the mutation rate is lower than 0.15, the algorithm is vulnerable to the quality of the initial solution and falls into local optima. The upper limit of 0.3 is set based on the need to maintain the stability of the algorithm. Exceeding this value will cause excessive destruction of high-quality gene characteristics and make the evolutionary process degenerate into a random search. Dynamically generate the mutation rate value before each mutation to increase population diversity and prevent the search trajectory from being homogenized.

[0054] For each individual in the first area, the gene locus of the material supply time on the chromosome is mutated with a probability of v. The mutation selection is not completely random, but the actual selection probability is adjusted according to the material importance to ensure that the gene locus of key medical equipment has a reasonable chance of mutation. By restricting the number of continuously selected gene loci, the over-concentration of mutations in specific regions of the chromosome is avoided, ensuring comprehensive coverage of the search space.

[0055] For the selected gene locus of the material supply time, a random offset of ±5 time units is made within the allowable range of the process dependency relationship. The offset range of [-5, 5] is the optimal value determined by simulating and analyzing the characteristics of the hospital engineering construction cycle, which can provide sufficient perturbation intensity while maintaining the feasibility of the solution.

[0056] The mutated individuals may violate the engineering constraints. The system adopts a two-stage repair technique to ensure the feasibility of the solution: by backtracking and querying in the process network model, the earliest feasible time point to satisfy the completion of the previous process is accurately calculated, and the supply time that violates the constraint is adjusted to this point. For the time points that violate the "peak shaving and valley filling" strategy is adopted. The supply volume is reduced according to the priority of the material importance, and the reduced amount is intelligently allocated to adjacent time points that do not violate the constraints.

[0057] Use the fitness function to evaluate the quality of the mutated individuals, and adopt a probabilistic acceptance mechanism: individuals with reduced fitness are accepted and retained with a probability of P1%. This soft selection mechanism can help the algorithm cross local sub-optimal regions. The value of P1 is dynamically adjusted according to the population diversity index. When the population diversity decreases, the acceptance probability is increased; otherwise, it is decreased, forming an adaptive balance. Preferably, P1 is taken as 20%.

[0058] As Figure 4 shown, the second area adopts a low-intensity mutation and elite retention strategy, aiming to perform fine optimization within the discovered high-quality solution space and improve the algorithm convergence efficiency. The individuals in the second area are sorted according to the fitness, and the optimal P2% individuals are directly selected to enter the next generation: the value of P2 is not fixed, but is dynamically adjusted according to the algorithm running stage and the concentration of solution quality. It is relatively low in the initial stage (about 10%) and increases to 20% - 30% with the increase of iterations. When selecting elite individuals, not only the absolute value of the fitness is considered, but also the quality difference between individuals is considered to ensure the diversity within the elite group.

[0059] For non-elite individuals, a low mutation rate is used for fine mutation: the lower mutation rate ensures fine-tuning of the existing high-quality solutions rather than large changes, which is suitable for local fine search. The actual mutation rate is adjusted according to the material type, and the gene locus of medical precision equipment obtains a lower mutation rate, while ordinary building materials obtain a relatively higher mutation rate.

[0060] Perform a small offset within the range of [-2, 2] on the selected gene positions: The small-range offset focuses on the fine exploration of the solution space and is suitable for finding the optimal solution within the known better solution area. The offset values are generated according to the normal distribution, making the probability of small offsets higher and enhancing the ability to accurately locate the optimal supply time point.

[0061] Conduct a fine constraint check and repair on the mutated solution: Not only check the completion time of the previous processes, but also consider the resource preparation time and process requirements to ensure that the repaired solution is feasible in engineering practice. When storing constraint violations, instead of adjusting a single time point, analyze the resource allocation of multiple adjacent time points to achieve constraint satisfaction with minimal interference. Through this two-region differential evolution strategy, the algorithm realizes the organic combination of global search and local optimization, and is particularly suitable for resource scheduling optimization problems in hospital engineering with multiple resource types and complex constraints.

[0062] This solution designs a unique two-region co-evolution mechanism, breaking through the limitations of the single-population evolution of traditional genetic algorithms. This mechanism realizes the dual evolution of the population by establishing a global search region (the first region) with a high mutation rate and a local optimization region (the second region) with elite retention, and through periodic individual migration and adaptive adjustment, ensures effective information exchange and characteristic complementarity between the two regions.

[0063] This mechanism is based on the "island model" theory in evolutionary computing, but has been innovatively adjusted according to the characteristics of the hospital engineering resource scheduling problem, introducing gradient diversity drive and regional adaptability transformation technologies. Through the "exploration - exploitation" balance mechanism, it solves the contradiction between the vast search space and the high optimization accuracy requirements in hospital engineering resource scheduling optimization.

[0064] The interval exchange period T is set to P3% (usually 5% - 10%) of the total number of iterations. This dynamic period mechanism is adaptively adjusted based on the problem scale, avoiding the problems of premature convergence or low search efficiency that may be caused by a fixed period. A too short period will lead to unstable regional characteristics, and a too long period will cause excessive regional differentiation. In hospital engineering optimization, a three-level period adjustment strategy is adopted according to the project complexity: Structural stage: a longer period (about 10%) to ensure sufficient exploration; Decoration stage: a medium period (about 7%) to balance exploration and exploitation; Equipment installation stage: a shorter period (about 5%) to focus on precise optimization.

[0065] The gradient diversity index is the core index to measure the coverage ability of individuals in the search space, and the calculation formula is: , where represents the supply time of the jth batch of the kth type of material in individual i, is the population average, is the material type weight. Select the top P4% (usually 5% - 8%) of individuals with the highest MD values as the diversity representatives. Although the fitness of these individuals may not be the highest, they can represent different search directions and provide multiple exploration paths for the algorithm.

[0066] The individuals in the second region already have high quality through elite selection. Select the top P5% (usually 3% - 5%) of individuals with the highest quality from the second region as the representatives of high-quality solutions. The design with P5% lower than P4% ensures that only the best solutions are introduced into the first region to avoid premature convergence.

[0067] Establish a two-way migration channel to migrate the diversity representatives in the first region to the second region and migrate the representatives of high-quality solutions in the second region to the first region, forming a dual mechanism of "diversity injection" and "sharing of high-quality solutions".

[0068] Adaptability adjustment of the migrated individuals in the first region: Reduce the random perturbation range from [-5, 5] to [-1, 1] so that the individuals with high diversity can stabilize after entering the fine optimization region and perform local fine search. Adaptability adjustment of the migrated individuals in the second region: Increase the mutation rate to v + 0.1 (usually reaching 0.25 - 0.4), and expand the perturbation range from [-2, 2] to [-7, 7] so that the high-quality solutions can serve as the search center in the first region and explore potential better solutions in a wider space.

[0069] After the migration is completed, conduct a complete re-evaluation of the populations in the two regions, update the fitness ranking, and dynamically adjust the selection pressure and mutation parameters for subsequent iterations according to the re-evaluation results to adapt to the changes in the population structure. This co-evolution mechanism realizes the dynamic balance between exploration and exploitation in the optimization of hospital engineering resources through the individual migration strategy driven by the gradient diversity index and the environmental adaptability adjustment after migration, and is particularly suitable for dealing with the collaborative optimization problem of ordinary building materials and precision medical equipment.

[0070] The algorithm sets three types of termination conditions: ① reaching the maximum number of iterations (usually 100 - 200 generations); ② the improvement of the optimal fitness is less than the threshold ε (usually 0.1%) for consecutive n generations (usually 20 generations); ③ finding a solution that meets the target fitness value (the fitness value is less than the preset target). Once any condition is met, the iteration is terminated. When the iteration terminates, extract the Pareto optimal solution set from the last generation of the population, apply the comprehensive scoring mechanism to the individuals in the Pareto optimal solution set, and calculate the weighted fitness value and select the individual with the highest comprehensive score as the optimal solution. The optimization quality of the processes related to special hospital areas (such as operating rooms and radiology departments) will be particularly considered in the scoring.

[0071] Adopt a hierarchical decoding strategy to decompose the chromosome into three levels: material type level, time point level, and supply quantity level. First, identify the material type code, then parse the supply time points of each type of material, and finally determine the specific supply quantity at each time point. Decode the supply time gene bits in the chromosome into the actual calendar time of the project. The system maps the relative time points to absolute dates according to the project start date and the working calendar (considering holidays, working hour limitations, etc.) to ensure that the scheduling plan corresponds to the actual construction schedule. According to the decoded supply time and quantity, the system makes intelligent batch divisions for each type of material. For ordinary building materials, the economic batch principle is adopted, and for medical special equipment, the installation timing requirements are given priority consideration.

[0072] The present application and its implementation manners are schematically described above. The description is not restrictive. Without departing from the spirit or basic features of the present application, the present application can be implemented in other specific forms. What is shown in the drawings is only one of the implementation manners of the present application, and the actual structure is not limited thereto. Therefore, if those of ordinary skill in the art are inspired by it and design similar structural manners and embodiments to this technical solution without creative efforts without departing from the purpose of this creation, they shall fall within the protection scope of the present application. In addition, the term "including" does not exclude other elements or steps, and the term "a" before an element does not exclude including "a plurality of" such elements. The terms such as "first" and "second" are used to indicate names and do not indicate any specific order.

Claims

1. A BIM-based intelligent construction site data processing method, characterized in that, Including: Obtain the basic data of the construction site using the BIM model to determine the construction resources to be optimized; The basic data includes design parameters, design construction period, and process dependency relationships; The construction resources to be optimized include material supply; Compare the real-time collected construction site data with the basic data obtained from the BIM model, and calculate the key deviation values, where the key deviation values include construction progress deviation and resource utilization rate deviation; Based on the key deviation values, establish a construction resource optimization model, where the construction resource optimization model aims to minimize the deviation values and sets construction constraint conditions according to the basic data; Use the improved genetic algorithm to solve the construction resource optimization model to obtain the optimal construction resource allocation plan, including: Encode the construction resource allocation plan into chromosomes, where the chromosomes contain material supply time and quantity; Construct a fitness function according to the multi-objective optimization function; Generate an initial solution set based on the BIM model as the initial population; Divide the population into the first area and the second area: In the first region, a mutation operation with a mutation rate greater than the threshold value is used to perform a global search; In the second region, a mutation rate less than the threshold value is adopted and an elitist retention strategy is used for local optimization; Set the co-evolution mechanism between the first area and the second area; When the iteration termination condition is met, output the optimal individual as the optimal construction resource allocation plan; The fitness function is as follows: , where is the weight coefficient; SE is the device sensitivity level; F is the value of the multi-objective optimization function; Among them, the improved genetic algorithm includes: Generate an initial solution set based on the BIM model, where the initial solution set satisfies the key construction constraints, and the key construction constraints include process logic constraints.

2. The BIM-based intelligent construction site data processing method according to claim 1, characterized in that: Calculating the key deviation values includes: Use the Internet of Things devices deployed at the construction site to collect real-time construction site data, where the real-time construction site data includes construction progress data and resource consumption data; Structurally process the real-time construction site data according to the component coding system of the BIM model to associate the spatial position relationship between the real-time construction site data and the BIM model basic data; Calculate the construction progress deviation value by comparing the actual construction progress with the BIM model design construction period; Calculate the resource utilization rate deviation value by comparing the material consumption with the BIM model design parameters.

3. The BIM-based intelligent construction site data processing method according to claim 1, characterized in that: Based on the key deviation values, establish a construction resource optimization model, including: Construct a multi-objective optimization function according to the construction progress deviation value and the resource utilization rate deviation value; Extract construction constraint conditions according to the basic data of the BIM model, where the construction constraint conditions include: material storage space constraint.

4. The BIM-based intelligent construction site data processing method according to claim 3, characterized in that: The multi-objective optimization function is: ； Among them, represents the construction progress deviation value of the \(i\)-th process, represents the resource utilization rate deviation value of the \(j\)-th type of material, and are weight coefficients, and satisfy ; The storage space constraint of materials is: , where represents the storage quantity of the k-th type of material at time t, represents the storage space occupied by a unit of the k-th type of material, represents the total available storage space at the construction site at time t.

5. The BIM-based intelligent construction site data processing method according to claim 1, characterized in that: Generate an initial solution set based on the BIM model, including: Extract the process dependency relationship and material demand information from the BIM model to construct a process network model in the form of a directed acyclic graph; According to the construction resources to be optimized, calculate the material types and quantities required for each process to form a material demand matrix M; According to the process network model and the material demand matrix M, generate an initial material supply plan through heuristic rules, where the initial supply plan contains the supply time points for each batch; Modify the initial supply plan to meet the process dependencies by moving the supply time point back and forth; Check whether the revised supply plan meets the material storage space constraint If not, revise the supply plan again until the storage space constraint is met at all time points; Encode the modified supply plan as chromosomes to form the initial population.

6. The BIM-based intelligent construction site data processing method according to claim 1, characterized in that: In the first region, a mutation operation with a mutation rate greater than the threshold is used for global search, including: Set the mutation rate threshold , and generate an actual mutation rate v within a preset range during each mutation operation; Randomly select the material supply time gene positions of the individuals in the first area of the population for mutation; Randomly offset the supply time of the selected gene positions within the allowable range of the obtained process dependencies; Conduct process logic constraint tests on the mutated individuals. For individuals that violate the constraints, determine valid time points through the process network for repair; Conduct material storage space constraint tests on the mutated individuals. For individuals that violate the constraints, balance by adjusting the material supply quantities at relevant time points; Use the fitness function to evaluate the fitness values of the mutated individuals, retain the individuals with improved fitness, and accept the individuals with decreased fitness with a preset probability to maintain population diversity.

7. The BIM-based intelligent construction site data processing method according to claim 1, characterized in that: In the second region, a mutation rate less than the threshold value is adopted and an elitist retention strategy is used for local optimization, including: Set the mutation rate threshold and generate a mutation rate within a preset range during each mutation operation; Sort the individuals in the second region of the population according to the fitness value, and select the individuals with the highest fitness value accounting for a preset proportion as the elite individuals to directly enter the next generation; ​ Select the material supply time gene positions of non-elite individuals for mutation; Offset the supply time of the selected gene positions within the allowable range of the process dependencies; Conduct process logic constraint tests on the mutated individuals. For individuals that violate the constraints, determine valid time points through the process network for repair; Conduct material storage space constraint tests on the mutated individuals. For situations that violate the constraints, balance by adjusting the material supply quantities at multiple time points; Use the fitness function to evaluate the fitness values of the mutated individuals, only retain the individuals with improved fitness, and eliminate the individuals with decreased fitness to enhance the local optimization efficiency.

8. The BIM-based intelligent construction site data processing method according to any one of claims 1 to 6, characterized in that: Set up a co-evolution mechanism between the first area and the second area, including: Set an interval exchange period and perform an interval exchange once every preset proportion of the iteration times; Select representative individuals from the first area based on the diversity evaluation index, and select high-quality solution individuals from the second area based on the fitness value; Execute two-way migration, migrate the representative individuals in the first area to the second area, and migrate the high-quality solution individuals in the second area to the first area; Conduct adaptive adjustment on the migrated individuals so that the individuals migrated from the first area to the second area adapt to the fine search environment, and so that the individuals migrated from the second area to the first area enhance the global exploration ability; Use the fitness function to evaluate the first area and the second area after migration and update the fitness ranking.

Images