Building construction process intelligent management method and system based on BIM technology

Abstract

The invention provides a building construction process intelligent management method and system based on a BIM technology, and relates to the technical field of construction process management, and the method comprises the steps: building a BIM model according to the current project information, dividing a construction plan into a plurality of links, and deploying the links into the model; simulating a construction process for multiple times through BIM software to obtain a transition probability; integrating resource reserve, construction duration and cost to construct a state space and an action space, and constructing an MDP model based on the state space and the action space; inputting the duration and the cost of the current construction link to the MDP model to predict the next construction link; by iteratively updating the current construction link and executing prediction until the stop condition is met, the intelligent and refined construction management is realized by combining the BIM model and the MDP model.

Description

Technical Field

[0001] The present application relates to the technical field of construction process management, and in particular to an intelligent management method and system for a building construction process based on BIM technology. Background Art

[0002] As the core carrier of digital transformation in the construction field, Building Information Modeling (BIM) technology realizes the integrated management and collaborative application of data throughout the life cycle of buildings by constructing parametric three-dimensional visualization models. This technology deeply integrates multi-dimensional data such as geometric information, physical properties and process parameters, and provides a data-driven technical support system for each stage of construction project planning, design, construction and operation and maintenance.

[0003] In the existing BIM-assisted construction process, there is a significant separation between the model development stage and the construction implementation stage: the technical team first completes the BIM construction drawing design, and then transfers the BIM model to the construction unit. The construction team needs to perform secondary information conversion based on the two-dimensional drawings, complete the hierarchical decomposition of the construction process through manual interpretation, and finally form a process chain to guide on-site operations. This process is essentially a linear transmission mode of "model development-information conversion-construction execution".

[0004] It is worth noting that the existing process planning methods are highly dependent on expert experience judgment and deterministic algorithms such as the critical path method (CPM), which are essentially static scheduling mechanisms. Since the dynamic mapping relationship of the construction scene has not been established, when faced with real-time disturbance factors such as resource reserves, construction costs, and construction time, it is easy to cause systemic problems such as process connection failure and resource conflicts, resulting in an increase in the probability of risks such as construction cost overruns and schedule delays. This extensive management model based on empiricism has been unable to adapt to the refined control needs of modern construction projects for multi-objective dynamic optimization. Summary of the invention

[0005] In order to dynamically predict the construction links and realize the intelligent and refined construction management process, the present application provides a method and system for intelligent management of the construction process based on BIM technology.

[0006] In the first aspect, the present application provides a method for intelligent management of building construction process based on BIM technology, which adopts the following technical solutions: An intelligent management method for a building construction process based on BIM technology comprises the following steps: First modeling: Establish a BIM model based on the information of the current project; First collection: obtain the construction plan of the current project, divide the construction plan into multiple construction links, deploy the construction links into the BIM model, and obtain a new BIM model; Construction simulation: Use BIM software to simulate the construction process multiple times in the new BIM model to obtain the transfer probability between each construction link; Constructing the MDP model: Integrate resource reserves, construction duration and construction costs of the construction phase into the state space, integrate the actions that can be taken in each state in the state space into the action space, and construct the MDP model based on the state space, action space and transition probability; Predict construction phase: obtain the current resource reserves, construction time and construction cost of completed construction phases, input the current resource reserves, construction time and construction cost of completed construction phases into the MDP model, and predict and output the next construction phase; Iteration: Record the construction sequence of the next construction link, update the next construction link to the current construction link, and execute the steps of predicting the construction link until all construction links are predicted.

[0007] This application converts project information into a three-dimensional digital form by constructing a BIM model, which helps the project team to better understand the project structure and layout, discover design conflicts and problems in advance, and thus reduce changes and errors in the construction process. Subsequently, this application deploys the construction links into the BIM model, and the project team can clearly see the arrangement of each construction link in a three-dimensional environment, thereby optimizing the construction sequence and space utilization. Subsequently, this application simulates the construction process multiple times through BIM software to obtain the transfer probability between construction links. Subsequently, this application constructs an MDP model to provide a scientific framework for construction decisions, taking into account the uncertainty and dynamics in the construction process. Subsequently, this application inputs the current resource reserves, the construction time and construction cost of the completed construction links into the MDP model, and outputs the next construction link to improve the scientificity and rationality of construction decisions. The combined use of the BIM model and the MDP model realizes the refined management and scientific decision-making of construction projects. The BIM model provides a detailed three-dimensional digital representation, and the MDP model provides a scientific framework for construction decisions. The combined use of precise modeling and visualization technology not only improves the scientificity and rationality of construction decisions, but also reduces changes and errors during the construction process, reduces project risks, and realizes the intelligence and refinement of the construction management process.

[0008] Optionally, after executing the step of building the MDP model and before executing the step of predicting the construction link, the method further includes: First setting: Set the reward function of the MDP model. The calculation model is as follows: ; ; ; ; in, is the reward function; is the weight of the construction cost of the tth construction link; is the expected construction cost of the tth construction stage; is the actual construction cost of the tth construction link; is the weight of the construction duration of the tth construction link; is the expected construction duration of the tth construction link; is the actual construction duration of the tth construction link; is the weight of the transfer probability of the tth construction link; From the tth construction link Take Action Then transfer to the next construction stage The transition probability of is the actual construction cost of the p-th action; is the actual construction time of the pth action; n is the time from the tth construction link To the next construction stage The number of actions taken.

[0009] Optionally, after executing the step of predicting the construction link and before executing the step of iterating, the method further includes: Constructing a matrix: constructing a reward matrix according to the reward function, wherein the element in the i-th row and the j-th column of the reward matrix represents the value of the reward function of the i-th construction link with respect to the j-th action; Predict construction actions: Based on the reward matrix, use the gradient descent method to determine the action sequence that can maximize the total reward; Deployment: deploy the action sequence to the corresponding construction link.

[0010] The present application constructs a reward matrix based on the reward function, in which the element in the i-th row and j-th column represents the reward value of the j-th action in the i-th construction link, and the reward matrix is ​​used to quantify the rewards of different actions for each action in different construction links. Subsequently, the present application uses the gradient descent method to determine the action sequence that can maximize the reward value based on the reward matrix. Finally, the present application deploys the determined action sequence to the corresponding construction link to guide the actual construction. The present application improves the scientificity and accuracy of the optimization process by using mathematical methods and optimization algorithms such as the reward matrix and the gradient descent method.

[0011] Optionally, after performing the iterative step, the method further comprises: Second collection: obtain the action sequence corresponding to each construction link, and build a Gantt chart of the construction link according to the action sequence corresponding to each construction link; The third collection: obtain the current construction progress, map the current construction progress to the Gantt chart of the construction link, and obtain the updated Gantt chart of the construction link; Deviation judgment: Determine whether there is a construction deviation based on the updated Gantt chart of the construction link. If yes, execute the alarm step; if no, execute the step of continuing construction; Alarm: send out an alarm signal; Continue construction: Continue construction according to the updated Gantt chart of the construction phase.

[0012] This application obtains the action sequence corresponding to each construction link, and constructs a Gantt chart of the construction link based on these action sequences to clearly show the timing and sequence of each action in the construction link. By constructing a Gantt chart, the project team can develop a detailed construction plan and clarify the specific actions and time nodes of each construction link. Subsequently, this application obtains the real-time construction progress and maps the current construction progress to the Gantt chart of the construction link to obtain an updated Gantt chart of the construction link, so that the Gantt chart can reflect the current construction status and provide real-time data support for deviation judgment.

[0013] Optionally, after the step of performing deviation determination and before the step of performing alarm, the method further includes: Add tags: add a completed or uncompleted tag to each construction link according to the current construction progress; Fourth collection: obtaining the construction link Gantt chart corresponding to the construction link under the unconstructed label, recorded as the unconstructed Gantt chart; Progress judgment: Based on the unconstructed Gantt chart, determine whether there are completed actions in the construction link under the unconstructed label. If so, execute the step of calculating the current progress; if not, execute the step of warning; Calculate the current progress: Use statistical analysis algorithms to calculate the current completion rate based on the Gantt chart of all construction links; Calculate the plan progress: Use statistical analysis algorithms to calculate the plan completion degree at the current moment based on the construction plan; Calculate the difference: Calculate the difference between the current completion and the planned completion; Re-judge: judge whether the difference meets the expectation, if so, execute the step of continuing construction; if not, execute the step of warning.

[0014] This application first adds a constructed label or an unconstructed label to each construction link according to the current construction progress. Subsequently, this application obtains the construction link Gantt chart corresponding to the construction link under the unconstructed label, which is recorded as the unconstructed Gantt chart. Subsequently, this application determines whether there is a completed action based on the unconstructed Gantt chart to obtain the actual progress of the construction. Subsequently, this application uses a statistical analysis algorithm to calculate the current completion degree to provide data support for progress management. Subsequently, this application calculates the planned completion degree at the current moment based on the construction plan, and compares it with the current completion degree to evaluate the compliance of the construction progress. Subsequently, this application calculates the difference between the current completion degree and the planned completion degree to identify the degree of deviation between the current construction progress and the planned construction progress. Determine whether the difference meets expectations. If so, execute the step of continuing construction; if not, execute the step of alarming. By adopting the above scheme, this application improves the scientificity, accuracy, flexibility and adaptability of construction progress management, and provides a strong guarantee for the smooth progress of the project.

[0015] Optionally, the reward function can also adopt the following calculation model: ; ; ; ; ; ; in, For Action The penalty factor of is an intermediate variable; is the reward function.

[0016] Optionally, after executing the step of predicting the construction link and before executing the step of iterating, the method further includes: Fifth collection: Obtain historical projects, take the project start date as the reference point, and convert the start time and end time of each construction link in the historical projects and / or the current project into a relative timestamp; Time alignment: Use the DTW algorithm to align each construction link of the historical project with each construction link of the current project to obtain the alignment result; Preconfiguration: preconfigure resources for each construction link of the current project according to the alignment results to obtain resource preconfiguration results; Resource judgment: Based on the resource pre-configuration results, determine whether the current resource reserves meet the needs of the next construction phase. If so, execute the iteration step; if not, execute the resource preparation step; Prepare resources: Increase resource reserves based on resource pre-configuration results.

[0017] This application first obtains historical projects, and uses the project start date as the reference point to convert the start time and end time of each construction link in the historical project and / or the current project into a relative timestamp, thereby achieving comparability of construction link time between different projects. Subsequently, this application uses the DTW (dynamic time warping) algorithm to align each construction link of the historical project with each construction link of the current project to obtain an alignment result. The DTW algorithm can process time series of different lengths and achieve flexible alignment of construction links. Subsequently, this application performs resource pre-configuration for each construction link of the current project based on the alignment results, so as to proactively plan the resources required for construction and minimize resource shortages or waste. Based on the experience of historical projects, this application can more reasonably configure the resources of the current project, improve resource utilization efficiency, and enable the construction process to proceed smoothly. According to the resource pre-configuration results, it is determined whether the current resource reserves meet the needs of the next construction link. Through resource judgment, the problem of insufficient resource reserves can be discovered in a timely manner, providing early warning for resource preparation. If the resources do not meet the demand, the current resource reserves are increased according to the resource pre-configuration results. By adopting the above scheme, this application improves the foresight, flexibility, efficiency and accuracy of project management, and provides a strong guarantee for the smooth progress of the project.

[0018] Optionally, in the time alignment step, the path search window R is set using the Sakoe-Chiba Band constraint algorithm, and the calculation model of the path search window is as follows: ; Among them, M is the number of construction links in the historical project; N is the number of construction links in the current project.

[0019] This application uses the Sakoe-Chiba Band constraint algorithm to set the path search window, which limits the range of the DTW algorithm when searching for the best path, so as to reduce the amount of calculation and improve the efficiency of the algorithm. Without constraints, the DTW algorithm may over-distort the path in order to match the two time series, resulting in unrealistic alignment results. The Sakoe-Chiba Band constraint reduces the occurrence of such over-distortion and improves the accuracy of the alignment by limiting the search range of the path.

[0020] In the second aspect, the present application provides an intelligent management system for the construction process based on BIM technology, which adopts the following technical solutions: An intelligent management system for building construction process based on BIM technology, comprising: a processor and a memory, The memory stores program code; When the processor calls the program code in the memory, the steps of the method described in the first aspect are executed.

[0021] In summary, the present application includes at least one of the following beneficial technical effects: 1. This application converts project information into a three-dimensional digital form by constructing a BIM model, which helps the project team to better understand the project structure and layout, discover design conflicts and problems in advance, and thus reduce changes and errors in the construction process. Subsequently, this application deploys the construction links into the BIM model, and the project team can clearly see the arrangement of each construction link in a three-dimensional environment, thereby optimizing the construction sequence and space utilization. Subsequently, this application simulates the construction process multiple times through BIM software to obtain the transfer probability between construction links. Subsequently, this application constructs an MDP model to provide a scientific framework for construction decisions, taking into account the uncertainty and dynamics of the construction process. Subsequently, this application inputs the current resource reserves, the construction time and construction costs of the completed construction links into the MDP model, and outputs the next construction link to improve the scientificity and rationality of construction decisions. The combined use of the BIM model and the MDP model realizes the refined management and scientific decision-making of construction projects. The BIM model provides a detailed three-dimensional digital representation, and the MDP model provides a scientific framework for construction decisions. The combined use of precise modeling and visualization technology not only improves the scientificity and rationality of construction decisions, but also reduces changes and errors during the construction process, reduces project risks, and realizes the intelligence and refinement of the construction management process.

[0022] 2. This application first obtains historical projects, and uses the project start date as the reference point to convert the start time and end time of each construction link in the historical project and / or the current project into a relative timestamp, thereby achieving comparability of construction link time between different projects. Subsequently, this application uses the DTW (Dynamic Time Warping) algorithm to align each construction link of the historical project with each construction link of the current project to obtain an alignment result. The DTW algorithm can process time series of different lengths and achieve flexible alignment of construction links. Subsequently, this application performs resource pre-configuration for each construction link of the current project based on the alignment results, so as to plan the resources required for construction in a forward-looking manner and minimize resource shortages or waste. Based on the experience of historical projects, this application can more reasonably configure the resources of the current project, improve resource utilization efficiency, and enable the construction process to proceed smoothly. According to the resource pre-configuration results, it is judged whether the current resource reserves meet the needs of the next construction link. Through resource judgment, the problem of insufficient resource reserves can be discovered in time, providing early warning for resource preparation. If the resources do not meet the demand, the current resource reserves are increased according to the resource pre-configuration results. By adopting the above solution, this application improves the foresight, flexibility, efficiency and accuracy of project management, and provides a strong guarantee for the smooth progress of the project. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is a flow chart of Example 1 of the present application; Figure 2 is a flow chart of Example 2 of the present application; Figure 3 This is a flow chart of Example 3 of the present application. DETAILED DESCRIPTION

[0024] The following combination Figures 1 to 3 This application is described in further detail.

[0025] Embodiment 1: This embodiment discloses an intelligent management method for a construction process based on BIM technology, referring to Figure 1 , the method includes: S11 first modeling, S12 first collection, S13 construction simulation, S14 building an MDP model, S15 predicting construction links and S16 iteration. In this embodiment, a BIM model is first established according to the current project information, and the construction plan is divided into multiple links and deployed into the model; the construction process is simulated multiple times by BIM software to obtain the transition probability; then, the resource reserve, construction time and cost are integrated to build the state space and action space, and the MDP model is built based on this and the transition probability; then, the current resource reserve, the duration and cost of the completed construction link are input into the MDP model to predict the next construction link; finally, the current construction link is updated through iteration and the prediction is executed until the stop condition is met. The process of this embodiment is as follows: S11 is the first modeling process. It establishes a BIM model based on the information of the current project and uses the BIM model as the digital basis for project construction management.

[0026] This step first collects all relevant information about the current project, including architectural design drawings, structural details, material specifications, construction specifications, etc. Then, using BIM software, this information is integrated into a three-dimensional, visual, and data-rich BIM model. The BIM model not only shows the appearance and internal structure of the building, but also includes the properties and relationships of the components and various parameters during the construction process.

[0027] S12 first collects detailed construction steps and time schedules from the project management team or construction plan, and divides these construction plans into multiple construction links with clear start and end points. In the BIM model, a visual display of the association between construction links and building components or areas is formed.

[0028] S13 construction simulation uses the simulation function of BIM software to simulate the construction process multiple times according to different construction sequences. During the simulation, the start and end time of each construction link and the conversion relationship between them are recorded. Through statistical analysis, the transfer probability between each construction link is calculated.

[0029] S14 constructs the MDP model and establishes the Markov decision process (MDP) model to describe the decision-making problems in the construction process.

[0030] The state space of the MDP model is constructed with resource reserves, construction time and construction cost of all construction links as dimensions.

[0031] Determine the actions that can be taken in each state, such as adjusting the order of each construction action in the construction link, increasing resource input, etc., to form an action space.

[0032] Combine the state space, action space and transition probability to construct the MDP model.

[0033] S15 predicts the construction phase, obtains the current resource reserves, the construction time and construction cost of the completed construction phase in real time, inputs these data into the MDP model as the current state, and predicts and outputs the next construction phase according to the decision rules of the MDP model.

[0034] S16 iterates, records the construction sequence of the next construction link, updates the predicted next construction link to the current construction link, executes S15 to predict the construction link, and obtains the new next construction link. Iterates continuously until the preset stop conditions are met, such as project completion, reaching the scheduled construction progress, etc.

[0035] This embodiment first establishes a BIM model based on the current project information, and obtains the construction plan, divides it into multiple links and deploys it in the model; then, uses BIM software to simulate the construction process multiple times to obtain the transition probability between construction links; then, integrates resource reserves, construction time, construction cost and transition probability to build an MDP model; then, by inputting the data of the current construction link into the MDP model, predicts and outputs the next construction link; finally, by iteratively updating the current construction link and executing the steps of predicting the construction link, until the project is completed or the scheduled construction progress is reached, real-time tracking and adjustment of the construction process can be achieved.

[0036] Example 2: Reference Figure 2 The difference between this embodiment and the first embodiment is that after executing S14 to construct the MDP model and before executing S15 to predict the construction link, it also includes: S21 first setting, setting the reward function of the MDP model, the calculation model of the reward function is as follows: ; ; ; ; in, is the reward function; is the weight of the construction cost of the tth construction link; is the expected construction cost of the tth construction stage; is the actual construction cost of the tth construction link; is the weight of the construction duration of the tth construction link; is the expected construction duration of the tth construction link; is the actual construction duration of the tth construction link; is the weight of the transfer probability of the tth construction link; From the tth construction link Take Action Then transfer to the next construction stage The transition probability of is the actual construction cost of the p-th action; is the actual construction time of the pth action; n is the time from the tth construction link To the next construction stage The number of actions taken.

[0037] The method for calculating the expected construction cost is already mature. This embodiment provides a detailed estimate of the cost of each construction link based on the detailed design drawings, bill of quantities and supplier quotations of the project.

[0038] The method for calculating the expected construction duration is also mature. This embodiment uses project management tools (such as Gantt charts and network diagrams) to plan the progress of the project and determine the expected start and end time of each construction link.

[0039] Then, the prediction and construction phase S15 is executed, and after the prediction and construction phase S15 is executed and before the iteration S16 is executed, the following steps are also included: S22 constructs a matrix, constructs a reward matrix according to the reward function, and the element in the i-th row and j-th column of the reward matrix represents the value of the reward function of the i-th construction link with respect to the j-th action.

[0040] For each construction link and action combination, it is necessary to calculate the reward function of each construction link under different actions. By constructing a reward matrix, it is possible to systematically analyze which actions have the most significant impact on the reward function of the entire project.

[0041] S23 predicts construction actions and regards the action selection problem of the construction link as an optimization problem, with the goal of maximizing the total reward of the entire project. Initially, a sequence of actions is randomly selected as the starting point. Then, for each construction link, the gradient of the reward function under the current action sequence is calculated according to the reward matrix. The action is adjusted in the opposite direction of the gradient, that is, the action that can increase the reward is selected, and the action sequence is gradually optimized. Repeat the above process until the stopping condition is met (such as the change in the reward function is less than a certain threshold, or the maximum number of iterations is reached) to optimize the construction plan and improve construction efficiency and cost-effectiveness.

[0042] S24 deployment integrates the optimized action sequence obtained by the gradient descent method to form the final construction plan, and deploys the optimized construction plan into the BIM model, corresponding to the construction links one by one. In this way, in the BIM model, the corresponding actions of each construction link, as well as the sequence and flow of the entire construction process, can be intuitively viewed.

[0043] Then, S16 iteration is performed, and after performing S16 iteration, the following steps are further included: S31 second collection, obtaining the action sequence corresponding to each construction link through the construction management system or related database.

[0044] This embodiment uses the collected action sequences to construct a Gantt chart for each construction link.

[0045] In the construction phase Gantt chart, the horizontal axis represents time and the vertical axis represents different construction phases. Each construction phase is represented by a bar, the length of the bar represents the estimated duration of the phase, and the segments within the bar represent different action sequences.

[0046] S32 third collection, through on-site monitoring, manual recording or automated data collection system, obtain real-time construction progress information. Map the collected real-time construction progress information to the Gantt chart of the construction link, generate a bar under the current construction progress, and obtain an updated Gantt chart of the construction link.

[0047] The bars in this step are different in color, transparency, and level from the bars generated in the second acquisition of S31. In this embodiment, the bars generated in the second acquisition of S31 are located at the bottom layer with a transparency of 100%; the bars generated in the third acquisition of S32 are located at the upper layer with a transparency of 80%. In other embodiments, the level and transparency of different bars can also be adjusted as needed.

[0048] By updating the Gantt chart of the construction phase, the construction progress can be tracked in real time, providing a basis for subsequent deviation judgment and decision-making.

[0049] S33 deviation judgment, compare the actual construction time of each construction link in the updated construction link Gantt chart with the planned construction time to determine whether there is a construction deviation in the current construction progress. If so, execute S34 to add a mark; if not, execute S42 to continue construction. Construction deviation is manifested as progress lag or progress advance.

[0050] S34 adds a mark, and adds a completed label or a non-constructed label to each construction link according to the current construction progress. The completed label indicates that all work of the link has been completed, and the non-constructed label indicates that the link has not yet started or has not yet been completed.

[0051] S35 fourth collection, filtering out the construction links under the unconstructed label from the updated construction link Gantt chart, and obtaining the corresponding Gantt chart.

[0052] S36 progress judgment, analyze the unconstructed Gantt chart to determine whether there are completed actions. If so, execute S37 to calculate the current progress; if not, execute S41 to issue an alarm.

[0053] S37 calculates the current progress, using a statistical analysis algorithm, and calculates the current degree of completion based on the Gantt charts of all construction links (including those that have been constructed and those that have not been constructed). The current degree of completion indicates the proportion of the project's completed workload to the total workload of the entire project.

[0054] S38 calculates the planned progress, also using a statistical analysis algorithm, but calculates the planned completion degree at the current moment based on the original construction plan. The planned completion degree indicates the proportion of the workload that should be completed according to the original plan to the total workload of the entire project.

[0055] S39 calculates the difference between the current completion and the planned completion to quantify the project progress deviation. A negative difference indicates a delay in progress, while a positive difference indicates an advance in progress.

[0056] S40 re-judges and compares the calculated difference with the expected deviation range to determine whether the difference meets expectations. If so, S42 is executed to continue construction; if not, S41 is executed to issue an alarm.

[0057] S41 alarm, sends an alarm signal.

[0058] S42 continues construction according to the updated Gantt chart of the construction phases.

[0059] This embodiment first obtains the action sequence of each construction link and constructs a Gantt chart of the construction link, then collects the construction progress in real time and updates the Gantt chart. Next, it is determined whether there is a construction deviation at present, and a constructed or unconstructed label is added to the construction link. For the unconstructed link, it is further determined whether there is a completed action, and the difference between the current completion degree and the planned completion degree is calculated. Depending on whether the difference meets expectations, it is decided whether to issue an alarm signal or continue construction according to the Gantt chart, so that the project can proceed smoothly and be completed on time.

[0060] Example 3: Reference Figure 3 The difference between this embodiment and embodiment 2 is that the reward function can also adopt the following calculation model: ; ; ; ; ; ; in, For Action The penalty factor of is an intermediate variable; is the reward function.

[0061] After executing the S15 prediction construction phase and before executing the S16 iteration, it also includes: S51 The fifth collection is to extract relevant data of historical projects from the project management system or database. These data should include the start and end time of each construction link, as well as other possible relevant information (such as resource consumption, workload, etc.).

[0062] Taking the project start date as the reference point, convert the start and end time of each construction link in the historical project and / or the current project into a relative timestamp. The relative timestamp is the time difference relative to the reference point (such as the start date), which can be in days, hours or minutes. By converting the timestamp, the time difference caused by different start dates between different projects is eliminated, making the historical project and the current project comparable in time.

[0063] S52 time alignment, DTW alignment of each construction link of the historical project with each construction link of the current project to find the best match between them and obtain the alignment result.

[0064] The path search window R is set using the Sakoe-Chiba Band constraint algorithm. The calculation model of the path search window is as follows: ; Among them, M is the number of construction links in the historical project; N is the number of construction links in the current project.

[0065] S53 pre-configuration: Analyze the results of DTW alignment to understand the resource consumption of each construction link in the historical project and their similarity with the corresponding links in the current project. According to the alignment results, pre-configure resources for each construction link of the current project, including determining the type, quantity and time schedule of the required resources. Through resource pre-configuration, sufficient resources can be available during the construction process of the current project, reducing the risk of construction delays or interruptions caused by resource shortages.

[0066] S54 resource judgment, check the resource reserve of the current project, including the purchased materials, arranged manpower, etc. According to the resource pre-configuration results, judge whether the current resource reserve meets the needs of the next construction link. If it does, continue to execute S16 iteration; if not, execute S55 to prepare resources.

[0067] S55 prepares resources, determines the gap between the current resource reserve and the pre-configured demand according to the result of the resource judgment, and increases the current resource reserve according to the gap.

[0068] This embodiment first obtains historical projects and converts the start and end times of the construction links into relative timestamps based on the start date. Then, the DTW algorithm is used and the path search window R constrained by Sakoe-Chiba Band is set to align the historical projects with the construction links of the current project. According to the alignment results, resources are pre-configured for the construction link of the current project, and it is determined whether the current resource reserves meet the needs of the next construction link. If not, the resource reserves are increased according to the pre-configuration results so that the construction can proceed smoothly.

[0069] Embodiment 4: This embodiment discloses an intelligent management system for building construction process based on BIM technology, the system comprising: a processor and a memory, The memory stores program code; The processor executes the steps of the method when calling the program code in the memory.

[0070] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. An intelligent management method for building construction process based on BIM technology, characterized in that: include: First modeling: Establish a BIM model based on the information of the current project; First collection: obtain the construction plan of the current project, divide the construction plan into multiple construction links, deploy the construction links into the BIM model, and obtain a new BIM model; Construction simulation: Use BIM software to simulate the construction process multiple times in the new BIM model to obtain the transfer probability between each construction link; Constructing the MDP model: Integrate resource reserves, construction duration and construction costs of the construction phase into the state space, integrate the actions that can be taken in each state in the state space into the action space, and construct the MDP model based on the state space, action space and transition probability; Predict construction phase: obtain the current resource reserves, construction time and construction cost of completed construction phases, input the current resource reserves, construction time and construction cost of completed construction phases into the MDP model, and predict and output the next construction phase; Iteration: Record the construction sequence of the next construction link, update the next construction link to the current construction link, and execute the steps of predicting the construction link until all construction links are predicted.

2. The intelligent management method for construction process based on BIM technology according to claim 1 is characterized in that: After executing the step of building the MDP model and before executing the step of predicting the construction link, it also includes: First setting: Set the reward function of the MDP model. The calculation model is as follows: ； ； ； ； in, is the reward function; is the weight of the construction cost of the tth construction link; is the expected construction cost of the tth construction stage; is the actual construction cost of the tth construction link; is the weight of the construction duration of the tth construction link; is the expected construction time of the tth construction link; is the actual construction duration of the tth construction link; is the weight of the transfer probability of the tth construction link; From the tth construction link Take Action Then transfer to the next construction stage The transition probability of is the actual construction cost of the p-th action; is the actual construction time of the pth action; n is the time from the tth construction link To the next construction stage The number of actions taken.

3. The intelligent management method for construction process based on BIM technology according to claim 2 is characterized in that: After executing the steps of the prediction construction phase and before executing the iteration step, it also includes: Constructing a matrix: constructing a reward matrix according to the reward function, wherein the element in the i-th row and the j-th column of the reward matrix represents the value of the reward function of the i-th construction link with respect to the j-th action; Predict construction actions: Based on the reward matrix, use the gradient descent method to determine the action sequence that can maximize the total reward; Deployment: deploy the action sequence to the corresponding construction link.

4. The intelligent management method for construction process based on BIM technology according to claim 3 is characterized in that: After performing the iteration step, the method further comprises: Second collection: obtain the action sequence corresponding to each construction link, and build a Gantt chart of the construction link according to the action sequence corresponding to each construction link; The third collection: obtain the current construction progress, map the current construction progress to the Gantt chart of the construction link, and obtain the updated Gantt chart of the construction link; Deviation judgment: Determine whether there is a construction deviation based on the updated Gantt chart of the construction link. If yes, execute the alarm step; if no, execute the step of continuing construction; Alarm: send out an alarm signal; Continue construction: Continue construction according to the updated Gantt chart of the construction phase.

5. The intelligent management method for building construction process based on BIM technology according to claim 4 is characterized in that: After executing the step of deviation determination and before executing the step of alarm, the method further includes: Add tags: add a completed or uncompleted tag to each construction link according to the current construction progress; Fourth collection: obtaining the construction link Gantt chart corresponding to the construction link under the unconstructed label, recorded as the unconstructed Gantt chart; Progress judgment: Based on the unconstructed Gantt chart, determine whether there are completed actions in the construction link under the unconstructed label. If so, execute the step of calculating the current progress; if not, execute the step of warning; Calculate the current progress: Use statistical analysis algorithms to calculate the current completion rate based on the Gantt chart of all construction links; Calculate the plan progress: Use statistical analysis algorithms to calculate the plan completion degree at the current moment based on the construction plan; Calculate the difference: Calculate the difference between the current completion and the planned completion; Re-judge: judge whether the difference meets the expectation, if so, execute the step of continuing the construction; if not, execute the step of issuing an alarm.

6. The intelligent management method for building construction process based on BIM technology according to claim 2 is characterized in that: The reward function can also adopt the following calculation model: ； ； ； ； ； ； in, For Action The penalty factor of is an intermediate variable; is the reward function.

7. The intelligent management method for building construction process based on BIM technology according to claim 6 is characterized in that: After executing the steps of the prediction construction phase, before executing the iteration step, the following steps are also included: Fifth collection: Obtain historical projects, take the project start date as the reference point, and convert the start time and end time of each construction link in the historical projects and / or the current project into a relative timestamp; Time alignment: Use the DTW algorithm to align each construction link of the historical project with each construction link of the current project to obtain the alignment result; Preconfiguration: preconfigure resources for each construction link of the current project according to the alignment results to obtain resource preconfiguration results; Resource judgment: Based on the resource pre-configuration results, determine whether the current resource reserves can meet the needs of the next construction phase. If so, execute the iteration step; if not, execute the resource preparation step; Prepare resources: Increase resource reserves based on resource pre-configuration results.

8. The intelligent management method for building construction process based on BIM technology according to claim 7 is characterized in that: In the time alignment step, the Sakoe-Chiba Band constraint algorithm is used to set the path search window B. The calculation model of the path search window is as follows: ； Among them, M is the number of construction links in the historical project; N is the number of construction links in the current project.

9. An intelligent management system for building construction process based on BIM technology, characterized in that: include: processor and memory, The memory stores program code; When the processor calls the program code in the memory, the steps of the method according to any one of claims 1 to 8 are executed.

Images