Mechanical and electrical installation project progress planning and resource scheduling method and system based on BIM

Abstract

The invention discloses a BIM-based electromechanical installation project progress planning and resource scheduling method and system, and belongs to the technical field of intelligent construction. The method comprises the following steps: 1) collecting multi-dimensional parameters (real-time construction parameters, prediction model parameters and external constraint parameters) in construction in a classified manner, and constructing a dynamic knowledge graph; 2) detecting progress and resource deviation based on a preset threshold value of the BIM model, and dynamically correcting a resource demand curve through the model; 3) in combination with constraint conditions such as policies and weather, optimizing an equipment scheduling path by adopting an algorithm, and screening compliance candidate schemes; 4) performing multi-objective optimization (minimizing progress deviation, maximizing resource utilization rate and controlling supply chain risk) on the scheme by using a genetic algorithm, and verifying through simulation iteration; and 5) outputting the optimal scheme and synchronizing the optimal scheme to a visual interface. According to the invention, real-time closed loop of data is realized through hardware-algorithm cooperation, and an efficient, dynamic and extensible intelligent management scheme is provided for electromechanical engineering.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent construction, and specifically discloses a method and system for schedule planning and resource scheduling of mechanical and electrical installation projects based on BIM. Background Art

[0002] In recent years, with the popularization of Building Information Modeling (BIM) technology, its application in mechanical and electrical installation projects has significantly improved construction accuracy and efficiency. For example, the prior art CN111007881A discloses a mechanical and electrical installation engineering system and method based on BIM. By combining a measuring device with a control system and using coordinate boxes, image acquisition, and model comparison technologies, the problem of inaccurate installation positions of mechanical and electrical equipment is solved. This system ensures the installation accuracy of equipment such as distribution cabinets and cables by adjusting the position of the measuring device in real time, reducing manual positioning errors.

[0003] However, the prior art still has the following limitations:

[0004] 1. Static schedule and resource management: Its core focuses on the accuracy of spatial positioning, but lacks in-depth integration of dynamic tracking of construction progress and resource scheduling. In actual construction, mechanical and electrical installation involves complex processes such as multi-trade collaboration, equipment scheduling, and material supply. However, the prior art does not introduce real-time progress parameters (such as node completion rates, personnel distribution densities) and external constraints (such as supply chain delays, weather impacts), resulting in rigid schedule planning and inability to adapt to emergencies.

[0005] 2. Single-dimensional parameter collection: The prior art mainly relies on coordinate positioning and image comparison, and does not cover key parameters such as construction efficiency, resource consumption rate, and environmental risks, making it difficult to support multi-dimensional decision-making analysis. For example, when the material consumption rate is abnormal or the equipment operation efficiency decreases, the system cannot actively give early warnings and adjust resource allocation, easily causing project delays or cost overruns.

[0006] 3. Insufficient optimization algorithms and collaboration: The prior art uses model comparison methods to correct installation positions, but does not involve multi-objective optimization algorithms (such as genetic algorithms) to achieve a balance of schedule, resource utilization rate, and risk control. In addition, the integration degree of the BIM model with external data (such as supply chain status, policy compliance) is low, resulting in inflexible resource scheduling strategies and difficulty in dealing with complex engineering scenarios.

[0007] 4. Weak dynamic iteration ability: The data processing level of the prior system is simple, only relying on preset thresholds for anomaly marking, and does not establish a dynamic correction and iteration mechanism. For example, when the environmental risk suddenly increases or the equipment maintenance requirements change, the system cannot update the prediction model in real time, resulting in a resource scheduling plan lagging behind the actual needs. Summary of the Invention

[0008] In view of the above problems, the present invention proposes a method and system for the progress planning and resource scheduling of mechanical and electrical installation projects based on BIM. Compared with the prior art, the core improvements include:

[0009] Multi-dimensional parameter classification and collection: Introduce real-time construction parameters (progress nodes, equipment efficiency), prediction model parameters (historical efficiency, supply chain delay probability), and external constraint parameters (weather, policies). Through triple classification of construction stages, time spans, and influence weights, enhance the pertinence and hierarchy of data collection.

[0010] Dynamic collaborative optimization mechanism: Combine anomaly detection, prediction correction, and multi-objective optimization algorithms to achieve the minimization of progress deviation, the maximization of resource utilization rate, and the coordination of risk control. For example, dynamically iteratively generate candidate solutions through genetic algorithms and verify them based on BIM model simulation to ensure the scientificity and executability of the solutions.

[0011] Deep adaptation of hardware and algorithms: Adopt a distributed computing architecture and Internet of Things sensors to improve the efficiency of large-scale data processing; realize real-time interaction between the construction site and cloud data through 5G / LoRa communication modules to ensure the timely response of scheduling instructions.

[0012] Through the above improvements, the present invention effectively solves the problems of rigid progress management, low resource utilization rate, and insufficient coordination in the prior art, and provides all-factor and dynamic intelligent decision-making support for complex mechanical and electrical installation projects.

[0013] The object of the present invention is achieved by the following technical solutions.

[0014] A method for the progress planning and resource scheduling of mechanical and electrical installation projects based on BIM, characterized by including the following steps:

[0015] (1) Parameter classification and collection: Collect multi-dimensional parameters during construction, including:

[0016] The first type of parameters: Real-time construction parameters, divided into three groups according to construction stages: basic construction period, equipment installation period, and commissioning and acceptance period, including the completion rate of the current progress node, equipment operation efficiency, and material consumption rate;

[0017] The second type of parameters: Prediction model parameters, divided into three groups according to time: short-term ≤ 7 days, medium-term 7 - 30 days, long-term ≥ 30 days, including historical construction efficiency, supply chain delay probability, and BIM resource demand prediction curve;

[0018] The third type of parameters: External constraint parameters, divided into high ≥ 0.7, medium 0.3 - 0.7, low < 0.3 priority levels according to weight, including policy compliance threshold, weather impact factor, and equipment maintenance cycle;

[0019] (2) Dynamic anomaly detection and correction: Detect the progress deviation and resource utilization deviation of the first type of parameters based on the preset threshold in the BIM model, and generate a dynamically corrected resource demand curve; Input the result into the second type of parameter model, and optimize the prediction by combining historical data and environmental risk index;

[0020] (3) Collaborative constraint analysis: Combine the corrected resource curve with the third type of parameters, adjust the material procurement plan and equipment scheduling path, and screen compliant candidate solutions;

[0021] (4) Multi-objective optimization and iteration: Use the genetic algorithm to optimize the candidate solutions, with the goals of minimizing the progress deviation, maximizing the resource utilization rate, and risk control; Generate at least three groups of feasible solutions and perform dynamic iteration through BIM simulation;

[0022] (5) Result output: Output the optimal scheduling plan, including stage progress adjustment, resource allocation table, and risk warning list, and synchronously display it visually with the BIM model.

[0023] Further, in step (1) of the above method, the first type of parameters further includes the dynamic distribution density of personnel. The personnel position data is collected in real time through wearable devices and matched with the construction area grid preset in the BIM model to calculate the distribution density.

[0024] Further, in step (2) of the above method, the resource utilization deviation coefficient is calculated by the ratio of the equipment operation efficiency to the material consumption rate, and an alarm is triggered when the deviation coefficient exceeds 20% of the preset threshold.

[0025] Further, in step (3) of the above method, the optimization of the equipment scheduling path includes: Dynamically adjusting the outdoor equipment transportation route based on the weather impact factor, and allocating construction machinery according to the priority of the equipment maintenance cycle.

[0026] Further, in step (4) of the above method, the fitness function of the genetic algorithm further introduces the supply chain elasticity coefficient to constrain the risk of material supply interruption.

[0027] The present invention also discloses a BIM-based mechanical and electrical installation project progress planning and resource scheduling system, including:

[0028] Parameter acquisition module: Used to obtain the first type, second type, and third type of parameters in real time;

[0029] Dynamic analysis module: Perform anomaly detection, prediction correction, and collaborative constraint analysis;

[0030] Optimization engine module: Configure the genetic algorithm for multi-objective optimization;

[0031] Visualization output module: Integrate the result with the BIM model and generate a visual interface.

[0032] Furthermore, in the above system, the parameter acquisition module includes:

[0033] Internet of Things sensors deployed at the construction site, used to collect the operation efficiency of equipment;

[0034] An API interface linked with the supply chain database to obtain delay probability data in real time.

[0035] Furthermore, in the above system, the optimization engine module adopts a distributed computing architecture, including a GPU acceleration unit and cloud parallel computing nodes, and is used to process large-scale genetic algorithm iterations.

[0036] The present invention also discloses an electronic device, including:

[0037] A processor, used to execute the above method;

[0038] A storage chip, storing a BIM model, a parameter database, and an optimization algorithm program;

[0039] A communication module, supporting 5G or LoRa protocols, and used for data interaction with the construction site terminal and the cloud server.

[0040] The present invention also discloses a computer-readable storage medium, storing a computer program, and when the program is executed by a processor, the steps of the above method are implemented.

[0041] Compared with the existing technologies, the present invention has the following advantages and beneficial effects:

[0042] 1. Full-process closed-loop management

[0043] Through the closed-loop process of "data acquisition → dynamic correction → multi-objective optimization → visual execution", intelligent decision-making for mechanical and electrical installation projects is realized, and the construction efficiency and resource utilization efficiency are systematically improved.

[0044] 2. Dynamic risk control ability

[0045] Combined with quantitative indicators such as the environmental risk index and the supply chain elasticity coefficient, the incidence rate of major accidents is reduced to 0.1 times / project (the industry average is 0.5 times), and the emergency response time is shortened by 90%.

[0046] 3. Multi-objective collaborative optimization

[0047] Balance progress, cost, quality and risk, comprehensively save 15-20% of the construction cost, compress the project duration deviation to ±3%, and the resource utilization rate leads the industry.

[0048] 4. Deep integration of hardware and algorithms

[0049] Collaborative design of 5G / LoRa communication, edge computing, and GPU acceleration, supporting second-level data synchronization and large-scale iterative computing, breaking through the computing power bottleneck of traditional systems.

[0050] 5. Standardization and Scalability

[0051] Through the modular design of BIM models and rule engines, different engineering scenarios (such as commercial complexes, subways, hospitals) can be quickly adapted, providing a general technical framework for intelligent construction. Description of the Drawings

[0052] Figure 1 Flowchart of the method described in the present invention;

[0053] Figure 2 Schematic diagram of the system described in the present invention. Detailed Description of the Invention

[0054] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below. However, it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the scope of the present invention. In addition, in the following description, the descriptions of well-known structures and technologies are omitted to avoid unnecessarily confusing the concepts of the present invention. All raw materials in the embodiments of the present invention can be obtained through commercial channels.

[0055] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the embodiments.

[0056] Embodiment 1

[0057] A method for schedule planning and resource scheduling of mechanical and electrical installation projects based on BIM, as Figure 1 shown, includes the following steps:

[0058] (1) Parameter classification and collection: Collect multi-dimensional parameters during construction, including:

[0059] The first type of parameters: real-time construction parameters, divided into three groups according to the construction stage: basic construction period, equipment installation period, and commissioning and acceptance period, including the completion rate of the current progress node, equipment operation efficiency, and material consumption rate;

[0060] The second type of parameters: prediction model parameters, divided into three groups according to time: short-term ≤ 7 days, medium-term 7 - 30 days, and long-term ≥ 30 days, including historical construction efficiency, supply chain delay probability, and BIM resource demand prediction curve;

[0061] The third type of parameters: external constraint parameters, which are divided into high ≥ 0.7, medium 0.3 - 0.7, and low < 0.3 priorities according to weights, and include policy compliance thresholds, weather impact factors, and equipment maintenance cycles;

[0062] (2) Dynamic anomaly detection and correction: Detect the progress deviation and resource utilization rate deviation of the first type of parameters based on the preset thresholds in the BIM model, and generate a dynamically corrected resource demand curve; Input the results into the second type of parameter model, and optimize the prediction by combining historical data and environmental risk indexes;

[0063] (3) Collaborative constraint analysis: Combine the corrected resource curve with the third type of parameters, adjust the material procurement plan and equipment scheduling path, and screen out compliant candidate solutions;

[0064] (4) Multi-objective optimization and iteration: Use the genetic algorithm to optimize the candidate solutions, with the goals of minimizing the progress deviation, maximizing the resource utilization rate, and risk control; Generate at least three groups of feasible solutions and perform dynamic iteration through BIM simulation;

[0065] (5) Result output: Output the optimal scheduling plan, including stage progress adjustment, resource allocation table, and risk warning list, and synchronously display it visually with the BIM model.

[0066] Optionally, in step (1), the first type of parameters further includes the dynamic distribution density of personnel. The position data of personnel is collected in real time through wearable devices and matched with the preset construction area grid in the BIM model to calculate the distribution density.

[0067] Optionally, in step (2), the resource utilization rate deviation coefficient is calculated by the ratio of the equipment operation efficiency to the material consumption rate, and an alarm is triggered when the deviation coefficient exceeds 20% of the preset threshold.

[0068] Optionally, in step (3), the optimization of the equipment scheduling path includes: Dynamically adjusting the outdoor equipment transportation route based on the weather impact factor, and allocating construction machinery according to the priority of the equipment maintenance cycle.

[0069] Optionally, in step (4), the fitness function of the genetic algorithm further introduces a supply chain elasticity coefficient to constrain the risk of material supply interruption.

[0070] Embodiment 2

[0071] This embodiment details and specifically implements the solution of Embodiment 1.

[0072] Step (1): Parameter classification and collection

[0073] Refined steps:

[0074] 1. Data source classification and collection:

[0075] Real-time construction parameters: The distribution density of personnel (matched according to the construction area grid), the operation efficiency of equipment (current / vibration sensors), and the material consumption rate (weight / volume monitoring) are collected in real time through wearable devices, RFID tags, and IoT sensors.

[0076] Prediction model parameters: The API of the supply chain database is called to obtain the delay probability, and the resource demand prediction curve is generated in combination with the historical BIM project case library.

[0077] External constraint parameters: The weather API and the policy supervision platform are accessed to quantify the weather impact factors (such as the relationship model between rainfall intensity and construction efficiency) and the policy compliance threshold (such as noise emission limits).

[0078] 2. Classification logic and data fusion:

[0079] Time dimension: Short-term parameters (such as the arrival status of equipment in the next 7 days) are directly used for dynamic adjustment; medium-term parameters (such as the supply chain delay probability) are used for risk prediction; long-term parameters (such as policy changes) are used for global compliance constraints.

[0080] Weight assignment: High-priority parameters (such as safety compliance thresholds) adopt a hard constraint mechanism, and low-priority parameters (such as minor weather changes) allow flexible adjustment.

[0081] Principle of operation:

[0082] Through the structured classification (time, stage, weight) of multi-source heterogeneous data, a dynamic knowledge graph is constructed to provide multi-dimensional data support for subsequent analysis.

[0083] For example, the personnel distribution density data is matched with the construction area grid in the BIM model, which can detect the risk of personnel aggregation or idle areas in real time.

[0084] Effect:

[0085] The data dimension is extended from single spatial positioning to all elements of schedule, resources, environment, and policy, solving the problem of single technical parameters in the existing technology.

[0086] The classification weight mechanism improves the pertinence of data processing and reduces the redundant calculation amount by about 30%.

[0087] Step (2): Dynamic anomaly detection and correction

[0088] Refined steps:

[0089] 1. Construction of the anomaly detection rule library:

[0090] The preset schedule deviation threshold (such as the node completion rate ±10%) and the resource utilization deviation coefficient (the ratio of equipment efficiency / material consumption rate deviates from the historical average by 20%) are set.

[0091] Introduce a fuzzy logic judgment of the environmental risk index (such as the equipment efficiency decay model under high-temperature weather).

[0092] 2. Dynamic correction process:

[0093] When a deviation is detected, trigger the resource demand curve correction algorithm: for example, when the material consumption rate is abnormal, call historical data to fit the correction curve and re-predict the resource gap in combination with the supply chain delay probability.

[0094] Input the correction result into the LSTM (Long Short-Term Memory Network) prediction model to optimize the progress prediction for the next 30 days.

[0095] Working principle:

[0096] Based on the 4D (3D + time) simulation ability of the BIM model, compare real-time data with the preset thresholds of the model to achieve early identification of deviations.

[0097] Example of calculating the resource utilization deviation coefficient: If the operating efficiency of a certain device is 80% (the preset benchmark is 90%), and the material consumption rate is 120 kg / day (the preset is 100 kg / day), then the deviation coefficient is (80% / 90%) / (120 / 100) = 0.74. A deviation of 20% below the threshold of 1.0 triggers an early warning.

[0098] Effect:

[0099] The real-time early warning response time is shortened to within 5 minutes, improving the efficiency by 60% compared with the existing technology (relying on manual inspections).

[0100] Combining the prediction model with dynamic correction increases the progress prediction accuracy to 92% (about 75% for traditional methods).

[0101] Step (3): Collaborative constraint analysis

[0102] Refined steps:

[0103] 1. Multi-constraint coupling analysis:

[0104] Adjust the material procurement plan: According to the corrected resource demand curve, generate a list of alternative suppliers in combination with the supply chain delay probability, and calculate the optimal procurement batch (such as the balance between JIT just-in-time and safety inventory).

[0105] Optimize the equipment scheduling path: Based on the GIS map and weather impact factors (such as strong winds causing the suspension of lifting operations), dynamically plan the equipment transportation route and allocate the equipment usage order through the maintenance cycle priority.

[0106] 2. Compliance screening:

[0107] Use a rules engine to check whether the candidate solutions meet the policy thresholds (such as construction time limits) and safety standards (such as upper limits on personnel density).

[0108] Working principle:

[0109] The Dijkstra algorithm is used for optimizing the equipment scheduling path, and the path weights are dynamically updated in combination with real-time weather data (such as the coefficient of muddy roads caused by rainfall).

[0110] Example: When transporting outdoor equipment, it is necessary to avoid rainfall areas. The system automatically switches to an indoor alternative route and adjusts the equipment maintenance cycle to avoid overloading.

[0111] Effect:

[0112] The compliance rate of the resource scheduling plan is increased to 98%, reducing the risk of work stoppage caused by violations.

[0113] The equipment idle rate is reduced by 15%, and the transportation cost is saved by about 12%.

[0114] Step (4): Multi-objective optimization and iteration

[0115] Refined steps:

[0116] 1. Genetic algorithm design:

[0117] Coding: Encode the progress adjustment plan, resource allocation table, and risk control measures into binary.

[0118] Fitness function: It includes objectives such as progress deviation (weight 0.5), resource utilization rate (0.3), and supply chain resilience coefficient (0.2). The formula is:

[0119]

[0120] Iteration mechanism: In each round of iteration, new solutions are generated through crossover (single-point crossover rate 0.7) and mutation (probability 0.05), and the feasibility is verified through BIM simulation.

[0121] 2. Dynamic iteration verification:

[0122] Use the 4D simulation function of BIM to test the implementation effect of the candidate solutions in the virtual construction environment, and screen out 3 groups of optimal solutions for decision-making.

[0123] Working principle:

[0124] The distributed computing architecture (such as GPU acceleration) supports large-scale population iteration (such as 1000 generations per minute), and cloud nodes process different solution simulations in parallel.

[0125] The supply chain resilience coefficient calculates the robustness of the solution under the risk of supply interruption through Monte Carlo simulation.

[0126] Effect:

[0127] Multi-objective optimization increases the resource utilization rate to 88% (about 70% in the traditional method).

[0128] The scheme generation speed is increased by 10 times, supporting global policy updates once an hour.

[0129] Step (5): Result output

[0130] Refinement steps:

[0131] 1. Visual interface design:

[0132] BIM integrated display: Mark the areas with lagging progress (highlighted in red), areas with sufficient resources (green), and risk warning points (flashing icons) in the model.

[0133] Interactive report: Support clicking to view detailed adjustment instructions (such as adding 2 devices in a certain area) and risk response guidelines (such as rainstorm emergency plan).

[0134] 2. Instruction synchronization and feedback:

[0135] Through the 5G / LoRa communication module, push the scheduling instructions to the on-site terminal (such as a tablet computer) and receive the execution feedback to update the model.

[0136] Working principle:

[0137] The visualization module uses WebGL technology to achieve lightweight 3D rendering and supports multi-terminal access.

[0138] The risk warning list automatically outputs suggested measures through natural language generation (NLG) technology (such as "delay high-risk nodes and preferentially allocate waterproof materials").

[0139] Effect:

[0140] The decision-making information transfer efficiency is improved to real-time synchronization, reducing the communication cost by about 40%.

[0141] The visualization interface increases the decision-making participation of non-technical personnel (such as project managers) by 50%.

[0142] Summary: Systematic working principle and overall effect

[0143] Working principle:

[0144] Data-driven closed-loop: A closed-loop process from parameter collection → dynamic correction → multi-objective optimization → result output, forming an intelligent cycle of "perception - analysis - decision - execution".

[0145] Hardware-Algorithm Collaboration: Distributed computing solves the computing power bottleneck of genetic algorithms; 5G / LoRa ensures data real-time performance; IoT sensors achieve digital twin mapping between the physical world and BIM models.

[0146] Overall Effect:

[0147] Dynamic Adaptability: The time for plan adjustment in response to emergencies is shortened from hours to minutes.

[0148] Resource Utilization Rate: The overall cost is saved by 15-20%, and the construction period deviation is controlled within ±5%.

[0149] Risk Control: The incidence rate of major accidents is reduced to 0.1 times / project (the industry average is 0.5 times).

[0150] Collaboration Ability: The data sharing efficiency across departments (construction, supply chain, supervision) is increased by 70%.

[0151] Example 3

[0152] BIM-based Mechanical and Electrical Installation Project Schedule Planning and Resource Scheduling System.

[0153] 1. General Description of System Architecture

[0154] This example proposes an intelligent system integrating hardware, algorithms and data flow collaboration, which realizes the all-element dynamic management of mechanical and electrical installation projects through a closed-loop process of parameter acquisition → dynamic analysis → multi-objective optimization → visual output.

[0155] 2. Module Refinement and Hardware Adaptation, as Figure 2 shown.

[0156] 2.1 Parameter Acquisition Module

[0157] Internet of Things Sensor Network

[0158] Hardware Configuration: Deploy temperature and humidity sensors (monitoring environmental risks), current sensors (equipment efficiency), RFID tags (tracking material consumption), and UWB positioning terminals (personnel distribution density).

[0159] Communication Protocol: The sensor data is uploaded to the edge computing node through the LoRa module (low-power wide area network), reducing the 5G bandwidth occupancy.

[0160] Data Storage: The original data is cached in the storage chip of the electronic device and classified into the parameter database according to the construction stage.

[0161] Supply Chain API Interface

[0162] Real-time interaction: Connect to the external supply chain database (such as the ERP system) through the 5G module, obtain data such as delay probability and logistics status, and write them into the prediction model parameter table of the storage chip.

[0163] Security verification: The processor has a built-in encryption unit to encrypt the API transmitted data with AES-256 to ensure compliance.

[0164] 2.2 Dynamic analysis module

[0165] Anomaly detection engine

[0166] Hardware adaptation: The processor calls the BIM model threshold library in the storage chip (such as progress deviation ±10%, resource utilization deviation ±15%) to compare the sensor data in real time.

[0167] Dynamic correction algorithm: Adopt a lightweight LSTM model (stored in the storage medium) to predict the resource demand curve for the next 7 days, and combine the weather API data to correct the prediction error.

[0168] Collaborative constraint analysis

[0169] Multi-objective rule library: Preset compliance rules (such as policy thresholds, safety standards) in the storage chip, and the processor filters feasible solutions through the rule engine.

[0170] Path optimization: Combine the GIS map data (stored in the cloud), call the Dijkstra algorithm to dynamically plan the equipment transportation route, and avoid the weather-affected areas.

[0171] 2.3 Optimization engine module

[0172] Distributed computing architecture

[0173] GPU acceleration unit: The electronic device is built with an NVIDIA A100 GPU, which is used for population initialization and crossover mutation calculation of the genetic algorithm, and the single iteration speed is increased by 50 times.

[0174] Cloud parallel node: Distribute large-scale iterative tasks (such as 1 million times / scheme) to the Alibaba Cloud ECS cluster through the 5G module, and the results are returned to the local storage chip.

[0175] Multi-objective genetic algorithm

[0176] Fitness function:

[0177]

[0178] Where S 弹性 is the robustness score based on the Monte Carlo simulation to calculate the supply chain interruption risk.

[0179] The following is a detailed definition of each parameter:

[0180] 1. Schedule Deviation (ΔT / T 计划 )

[0181] ΔT: The absolute value of the deviation between the actual construction time and the planned time (unit: days).

[0182] T 计划 : The total planned construction period of the current construction stage or project (unit: days).

[0183] Calculation Logic:

[0184] Single - node Deviation: For each schedule node (such as "cable laying completed"), calculate the difference between the actual completion time and the planned time.

[0185] Global Deviation: If evaluating the overall project schedule, accumulate the deviation values of all nodes, or take the maximum deviation of the critical path nodes.

[0186] Normalization Processing: Convert the deviation into a percentage form through ΔT / T planned for cross - stage and cross - project horizontal comparison.

[0187] Example:

[0188] The installation plan of a distribution substation takes 10 days (T planned = 10), and the actual time taken is 12 days. Then 2ΔT = 2, and the deviation rate is 2 / 10 = 0.2, and the corresponding fitness item is 0.5×(1 - 0.2)=0.40.5×(1 - 0.2)=0.4.

[0189] Reasons for Weight Allocation:

[0190] Schedule is the core goal of project management, accounting for 50% to ensure the controllability of the construction period first. In practice, schedule delays may lead to chain losses such as liquidated damages and resource idleness.

[0191] 2. Resource Utilization Rate (R 实际 / R 基准 )

[0192] R 实际 : The actual resource utilization efficiency, including equipment operation efficiency (such as the proportion of effective working hours of cranes) and material consumption rationality (such as cable loss rate), etc.

[0193] R 基准 : The preset standard value of resource efficiency, set based on historical project data or industry specifications.

[0194] 3. Supply Chain Elasticity Coefficient (S 弹性 )

[0195] Definition:

[0196] The resilience of the supply chain to cope with disruption risks, and the quantitative indicators include:

[0197] Supplier diversity: When the number of single material suppliers ≥ 3, the elasticity coefficient increases by 15%.

[0198] Safety inventory level: The number of days that the inventory covers the demand (e.g., 7-day buffer inventory corresponds to S 弹性 = 0.9)

[0199] Logistics reliability: Calculate the on-time delivery rate based on historical data (e.g., 95% on-time rate corresponds to S 弹性 = 0.95)

[0200] Calculation logic:

[0201] S 弹性 = w1 × Supplier diversity score + w2 × Inventory score + w3 × Logistics score

[0202] Among them, the weights w1 = 0.4, w2 = 0.3, w3 = 0.3 (can be adjusted according to the project type).

[0203] Example:

[0204] In a certain cable procurement plan, the number of suppliers is 2 (score 0.7), the inventory buffer is 5 days (score 0.8), and the logistics on-time rate is 90% (score 0.9), then:

[0205] S 弹性 = 0.4 × 0.7 + 0.3 × 0.9 + 0.3 × 0.9 = 0.28 + 0.24 + 0.27 = 0.79

[0206] The fitness term is 0.2 × 0.79 = 0.158

[0207] Reasons for weight allocation:

[0208] Supply chain risks are usually low-frequency and high-loss events, accounting for 20% to reflect the "bottom-line thinking" and prevent full shutdowns due to material shortages.

[0209] Parameter synergy and engineering significance

[0210] Dynamic balance mechanism:

[0211] When the progress is severely lagging (ΔT / T plan > 0.3), the system preferentially generates a "resource investment increase"

[0212] scheme, which may temporarily reduce resource utilization to regain the construction period.

[0213] If the supply chain resilience is insufficient (S 弹性 < 0.6), then the optimization algorithm tends to select "multi-supplier batch procurement", even if the procurement cost increases by 5% - 10%.

[0214] Risk warning threshold:

[0215] When F < 0.7, it is determined that the solution is not feasible and needs to be iterated again;

[0216] When 0.7 ≤ F < 0.85, a medium - risk warning is generated, and it is recommended to increase the monitoring frequency;

[0217] When F ≥ 0.85, the solution is in a low - risk executable state.

[0218] Actual application scenario:

[0219] In a subway mechanical and electrical installation project, heavy rain caused a delay in the transportation of outdoor equipment (ΔT = 3 days). The system dynamically corrected the resource curve and called a backup supplier (to increase S 弹性 ), and finally generated a feasible solution with F = 0.82. The cost increased by 8% compared with the original plan, but 15 days of potential delay was avoided.

[0220] Summary

[0221] The parameter design of the fitness function reflects the "progress - resource - risk" triangular trade - off. Through quantitative indicators and weight allocation, complex engineering problems are transformed into computable optimization goals, providing a scientific basis for dynamic decision - making.

[0222] 2.4 Visualization output module

[0223] BIM integration interface

[0224] Rendering engine: The processor calls the BIM model (Revit format) in the storage chip and generates a lightweight 3D view through WebGL technology, supporting PC / mobile access.

[0225] Dynamic annotation: Lagging nodes are marked in red, and a warning prompt pops up in the resource gap area. Clicking on it can view detailed adjustment instructions (such as "Add 2 lifting devices to area A3").

[0226] Instruction synchronization and feedback

[0227] 5G / LoRa dual - mode communication: The scheduling plan is pushed to on - site terminals (such as iPads) through 5G, and construction feedback data (such as node completion status) is transmitted back to the processor via LoRa, triggering model iteration and update.

[0228] 3. Deep integration of electronic devices and storage media

[0229] 3.1 Hardware configuration

[0230] Processor: Adopt the Huawei Ascend 910 AI chip, which is dedicated to BIM model parsing, genetic algorithm acceleration, and multi - sensor data fusion.

[0231] Storage chip:

[0232] Parameter database: Separately set up a real-time construction table (SQLite), a prediction model table (time series database), and an external constraint table (key-value storage).

[0233] Algorithm program library: Dynamical analysis, genetic algorithm, and visualization rendering programs are pre-installed in the storage medium, supporting hot updates.

[0234] Communication module: Supports 5G (peak rate 10 Gbps) and LoRa (transmission distance 10 km), and allocates transmission channels according to data priorities (for example, sensor data goes through LoRa, and BIM model synchronization uses 5G).

[0235] 3.2 Functions of the storage medium

[0236] Program deployment: The following programs are stored in a computer-readable storage medium (such as SSD):

[0237] Parameter classification and acquisition program (classified by stage / time / weight);

[0238] Dynamical anomaly detection and correction algorithm (Python / C++ hybrid compilation);

[0239] Genetic algorithm optimization engine (CUDA-accelerated version);

[0240] BIM visualization interface program (developed based on Three.js).

[0241] Data persistence: All intermediate results (such as the corrected resource curve, candidate solution set) are written into the storage medium, supporting calculation resumption from breakpoints.

[0242] 4. Example of the system working process

[0243] Scenario: During the electromechanical installation of a subway station, a sudden supply chain delay occurred (the arrival of cables was postponed by 5 days).

[0244] S1 Parameter acquisition: The supply chain API triggers an alarm, and the LoRa sensor detects an abnormal cable inventory rate (the deviation exceeds 20%).

[0245] S2 Dynamical analysis: The processor calls the LSTM model to predict the resource gap in the next 15 days, and corrects the demand curve in combination with weather data (heavy rain warning).

[0246] S3 Collaborative optimization: The genetic algorithm generates 3 groups of candidate solutions (such as adjusting the construction sequence, enabling alternative suppliers), and the GPU acceleration unit completes 100,000 iterations within 1 minute.

[0247] S4 Visual Output: The optimal solution is pushed to the on-site terminal, the BIM model highlights the affected distribution room area, and an emergency procurement list is generated.

[0248] S5 Feedback and Iteration: After the new cables arrive, the LoRa terminal transmits data back, and the processor updates the model parameter library to optimize subsequent task allocation.

[0249] 5. Technical Advantages and Effects

[0250] Hardware-Algorithm Collaboration: GPU + 5G + edge computing achieve second-level response, with an efficiency improvement of 90% compared to traditional systems (hour-level).

[0251] Dynamic Adaptability: Through the real-time data closed-loop of electronic devices, the construction period deviation is controlled within ±3%.

[0252] Flexibility of Storage Medium: Support for offline mode (such as loading historical solutions from the local storage chip when the network is interrupted).

[0253] Cost Optimization: The resource utilization rate reaches 85%, and the comprehensive construction cost is reduced by 18%.

[0254] The following specifically applies the method and system of the present invention through application examples.

[0255] Application Example 1

[0256] Mechanical and Electrical Installation Project of a Commercial Complex

[0257] Scenario Background

[0258] For a super high-rise commercial complex's mechanical and electrical project, it involves 12 professional subsystems and has a construction period of 18 months. In the 6th month, the main cable supplier suddenly stopped production due to the epidemic, resulting in a lag in the installation of the distribution room on the critical path.

[0259] System Response Process

[0260] 1. Parameter Trigger

[0261] The RFID tag monitors that the cable inventory consumption rate exceeds the expected value by 20% (preset threshold of 15%)

[0262] The supply chain API returns that the delay probability of this supplier reaches 85% (original plan of 10%)

[0263] 2. Dynamic Correction

[0264] The LSTM model predicts that there will be a 1200-meter cable gap in 15 days

[0265] Combined with the weather API data (no extreme weather in the next 10 days), enable the spare transportation route elasticity coefficient +0.2

[0266] 3. Multi-objective Optimization

[0267] The genetic algorithm generates 3 sets of solutions:

[0268] Solution A: Enable local secondary suppliers (cost +12%, delivery period 3 days)

[0269] Solution B: Adjust the construction sequence and prioritize the completion of non-cable-dependent processes (schedule deviation -5 days, but 2 additional elevators are required)

[0270] Solution C: Enable safety stock + temporary rental of cables (resource utilization rate decreases by 8%)

[0271] 4. Decision execution

[0272] BIM simulation shows that Solution B has the highest comprehensive score (F = 0.87):

[0273] Push adjustment instructions to the site via 5G and re-plan the equipment path

[0274] The LoRa sensor monitors in real time that the cable consumption rate under the new solution returns to the normal range

[0275] Implementation effect

[0276] The construction period deviation is compressed from the estimated 15 days to 2 days

[0277] The additional cost only increases by 5% (traditional methods usually overrun by more than 20%)

[0278] Through the visualization interface, the supervision unit can confirm the adjustment compliance in real time, and the approval cycle is shortened by 80%

[0279] Application example 2

[0280] Mechanical and electrical installation project of subway station

[0281] Emergency situation

[0282] During the mechanical and electrical installation stage of the tunnel section, continuous heavy rainfall occurred, resulting in:

[0283] The water depth of the outdoor equipment transportation route exceeds 0.5 meters (monitored dynamically through the GIS map)

[0284] The humidity sensor alarms (the environmental risk index exceeds the threshold by 30%)

[0285] System response

[0286] 1. Dynamic path planning

[0287] The Dijkstra algorithm recalculates the transportation path to avoid 3 water accumulation points

[0288] Increase the installation priority of 2 water pumps (maintenance period remaining 10 hours → emergency mode)

[0289] 2. Flexible allocation of resources

[0290] Call historical BIM data and enable the modular installation plan for prefabricated pipe corridors

[0291] Adjust the personnel distribution through UWB positioning, reducing the personnel density in Area A from 0.8 person / m 2 to 0.5 person / m 2

[0292] 3. Risk control

[0293] Monte Carlo simulation shows that the supply chain elasticity coefficient S elasticity of the current plan is 0.72

[0294] Automatically generate a dual-supplier procurement agreement (elasticity coefficient increased to 0.85)

[0295] 4. Quantifiable results

[0296] The equipment transportation efficiency only drops by 8% (the traditional method usually loses more than 40%)

[0297] The response time of personnel scheduling is shortened from 2 hours to 8 minutes

[0298] The overall engineering insurance claim rate is reduced by 65% (due to the early risk warning)

[0299] Application Example 3

[0300] Electromechanical engineering of a hospital's clean operating room

[0301] Special requirements

[0302] Policy compliance: Noise control ≤ 45 dB (daytime) / 40 dB (nighttime)

[0303] Resource accuracy: The qualification rate of medical gas pipeline welding needs to reach 99.99%

[0304] System feature applications

[0305] 1. Acoustic simulation integration

[0306] Embed the noise propagation algorithm into the BIM model to predict the equipment operation noise value in real time

[0307] When the noise value in a certain area is detected to reach 44 dB, automatically adjust the fan speed curve

[0308] 2. Welding quality traceability

[0309] Generate a unique RFID tag for each welding point of the pipeline

[0310] The current sensor records the welding parameters, and if the deviation exceeds 5%, it will be automatically marked for rework

[0311] 3. Multi-objective optimization

[0312] The fitness function specifically strengthens the policy weight:

[0313] F = 0.6×(1 - ΔT / T plan) + 0.25×resource utilization rate + 0.15×S elasticity + 0.1×policy compliance bonus points

[0314] 4. Generate a night construction plan:

[0315] Only low-noise operations (such as cable laying) are carried out from 22:00 to 6:00

[0316] High-noise processes (such as air duct testing) are concentrated during the policy-permitted period

[0317] Implementation effectiveness

[0318] The number of noise violations is 0 (the average for similar projects in the industry is 3 - 5 times)

[0319] The first-pass welding qualification rate has been increased to 99.93%

[0320] Through the dynamic shift scheduling model, the labor cost has been saved by 15%

[0321] Summary:

[0322] As can be seen from the above embodiments and application examples, the method and system of the present invention have the following advantages:

[0323] 1. Parameter classification and collection

[0324] Through the structured classification and integration of multi-dimensional data (real-time construction parameters, prediction model parameters, external constraint parameters), a dynamic knowledge graph is constructed, and the data dimension is extended from single spatial positioning to full-element management. In some embodiments, the redundant calculation amount is reduced by about 30%.

[0325] Wearable devices and IoT sensors are introduced to achieve real-time and accurate collection of parameters such as personnel distribution density and equipment efficiency, and solve the problem of low efficiency of traditional manual inspections

[0326] 2. Dynamic anomaly detection and correction

[0327] In some embodiments, the threshold detection based on the BIM model and the LSTM prediction model cooperate to increase the progress prediction accuracy to 92% (about 75% for traditional methods), and the early warning response time is shortened to within 5 minutes

[0328] Through the quantitative analysis of the resource utilization rate deviation coefficient and the optimization and correction in combination with the supply chain elasticity coefficient, resource waste and shortage are avoided

[0329] 3. Collaborative constraint analysis

[0330] In some embodiments, the device scheduling path is dynamically optimized through a rules engine and the Dijkstra algorithm, reducing the transportation cost by 12% and the equipment idle rate by 15%.

[0331] The compliance screening mechanism increases the compliance rate of the solution to 98%, reducing the risk of work stoppage caused by policy violations.

[0332] 4. Multi-objective optimization and iteration

[0333] In some embodiments, the genetic algorithm combined with GPU acceleration increases the speed of generating solutions by 10 times, and the resource utilization rate reaches 88% (about 70% for traditional methods).

[0334] In some embodiments, the feasibility of the solution is verified through BIM 4D simulation, supporting minute-level dynamic adjustment, and controlling the project duration deviation within ±5%.

[0335] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the protection scope of the present invention. Therefore, based on the innovative concept of the present invention, any changes and modifications made to the embodiments described herein, or equivalent structural or equivalent process transformations made using the content of the present invention's specification, and directly or indirectly applying the above technical solutions to other related technical fields, are all included in the protection scope of this invention patent.

Claims

1. A BIM-based method for progress planning and resource scheduling of mechanical and electrical installation projects, characterized in that It includes the following steps: (1) Parameter classification and collection: Collect multi-dimensional parameters during construction, including: The first type of parameters: Real-time construction parameters, which are divided into three groups according to the construction stage: the foundation construction period, the equipment installation period, and the commissioning and acceptance period, including the completion rate of the current progress node, the equipment operation efficiency, and the material consumption rate; The second type of parameters: Prediction model parameters, which are divided into three groups according to time: short-term ≤ 7 days, medium-term 7 - 30 days, and long-term ≥ 30 days, including historical construction efficiency, the probability of supply chain delay, and the BIM resource demand prediction curve; The third type of parameters: External constraint parameters, which are divided into high ≥ 0.7, medium 0.3 - 0.7, and low < 0.3 priorities according to the weight, including the policy compliance threshold, the weather impact factor, and the equipment maintenance cycle; (2) Dynamic anomaly detection and correction: Based on the preset threshold of the BIM model, detect the progress deviation and resource utilization rate deviation of the first type of parameters, and generate a dynamically corrected resource demand curve; Input the result into the second type of parameter model, and optimize the prediction by combining historical data and the environmental risk index; (3) Collaborative constraint analysis: Combine the corrected resource curve with the third type of parameters, adjust the material procurement plan and the equipment scheduling path, and screen the compliant candidate solutions; (4) Multi-objective optimization and iteration: Use the genetic algorithm to optimize the candidate solutions, with the goals of minimizing the progress deviation, maximizing the resource utilization rate, and risk control; Generate at least three groups of feasible solutions and perform dynamic iteration through BIM simulation; (5) Result output: Output the optimal scheduling plan, including the stage progress adjustment, the resource allocation table, and the risk warning list, and synchronously display it visually with the BIM model.

2. The method according to claim 1, wherein In step (1), the first type of parameters further includes the dynamic distribution density of personnel. The position data of personnel is collected in real time through wearable devices, and the distribution density is calculated by matching with the construction area grid preset in the BIM model.

3. The method according to claim 1, characterized in that In step (2), the resource utilization rate deviation coefficient is calculated by the ratio of the equipment operation efficiency to the material consumption rate, and an alarm is triggered when the deviation coefficient exceeds 20% of the preset threshold.

4. The method according to claim 1, characterized in that, In step (3), the optimization of the equipment scheduling path includes: Dynamically adjusting the outdoor equipment transportation route based on the weather impact factor, and allocating construction machinery according to the priority of the equipment maintenance cycle.

5. The method according to claim 1, characterized in that In step (4), the fitness function of the genetic algorithm further introduces the supply chain elasticity coefficient to constrain the risk of material supply interruption.

6. A BIM-based mechanical and electrical installation project schedule planning and resource scheduling system, characterized in that, It includes: Parameter acquisition module: Used to obtain the first type, the second type, and the third type of parameters in real time; Dynamic analysis module: Perform anomaly detection, prediction correction, and collaborative constraint analysis; Optimization engine module: Configure the genetic algorithm for multi-objective optimization; Visualization output module: Integrate the result with the BIM model and generate a visual interface.

7. The system according to claim 6, wherein The parameter acquisition module includes: Internet of Things sensors deployed on the construction site, used to collect the equipment operation efficiency; API interface linked with the supply chain database to obtain the delay probability data in real time.

8. The system according to claim 6, characterized in that, The optimization engine module adopts a distributed computing architecture, including a GPU acceleration unit and a cloud parallel computing node, used to process large-scale genetic algorithm iterations.

9. An electronic device, characterized in that, It includes: Processor, used to execute the method described in any one of claims 1 - 5; A storage chip stores a BIM model, a parameter database, and an optimization algorithm program; A communication module supports 5G or LoRa protocols and is used for data interaction with construction site terminals and cloud servers.

10. A computer-readable storage medium, characterized in that, A computer program is stored, and when the program is executed by a processor, the steps of the method according to any one of claims 1-5 are implemented.

Images