A distributed scheduling and storage optimization method for engineering resources based on access popularity

Abstract

The present invention provides a method for distributed scheduling and storage optimization of engineering resources based on access heat, which relates to the technical field of building resource scheduling and management. The method comprises the following steps: obtaining engineering site data, identifying the current construction stage based on the site data, wherein the construction stage includes the main structure stage, the secondary structure stage and the decoration installation stage, and calculating the dynamic heat of each component, namely the access heat; obtaining map information of the construction site, and configuring a warehouse partition topology according to the current construction stage and the access heat of engineering resources; receiving data of all scheduling nodes in S2 and the partition results of the construction site in S3, and dynamically selecting a scheduling mode for engineering resources in combination with the current construction stage type and the access heat of engineering resources; monitoring whether there are any abnormal construction events based on the engineering site data in S1; and accurately allocating resources according to actual construction needs to avoid idle resources and waste, thereby improving the utilization efficiency of engineering resources.

Description

Technical Field

[0001] The present invention relates to the technical field of construction resource scheduling and management, and in particular to a method for distributed scheduling and storage optimization of engineering resources based on access popularity. Background Art

[0002] Currently, large-scale construction projects are huge in scale. With the continuous expansion of project scale and the increasing construction requirements, resource management faces many severe challenges. On the one hand, the construction schedule is often highly tight, the project cycle is long, and each link is closely linked. Any delay in resource allocation may lead to delays in the construction period and bring huge economic losses. On the other hand, the resources involved in construction projects are of various types and extremely dispersed, covering various aspects such as construction equipment, building materials, site resources, and software resources. Prefabricated components are large in size, heavy in weight, have a long production cycle, high transportation requirements, large storage space, and certain environmental requirements. They are often used in large and complex construction projects, and the timeliness and accuracy of their supply have a significant impact on the construction schedule.

[0003] The demand for prefabricated components in different construction stages of large and complex building projects shows a clear phased trend. During the main structure construction stage of super high-rise buildings, the demand for prefabricated components increases sharply, while during the secondary structure construction stage, the demand for prefabricated partition panels, prefabricated stairs, etc. becomes the focus.

[0004] At the same time, since the construction progress of large-scale projects is affected by many factors, the demand for prefabricated components and the timing of demand will also fluctuate greatly, increasing the difficulty of resource scheduling and storage;

[0005] Resource access demand exhibits highly dynamic characteristics. The frequency and duration of access to various resources are difficult to predict during different construction phases, seasons, and even weather conditions. For example, construction equipment usage is concentrated during specific construction phases, and the access frequency of building materials is affected by construction progress and usage. Traditional resource management methods struggle to cope with such complex and changing resource access patterns, often leading to the coexistence of idle resources and resource shortages. Summary of the Invention

[0006] In view of the shortcomings of the existing technology, the present invention provides an access popularity-based distributed scheduling and storage optimization method for engineering resources.

[0007] A method for distributed scheduling and storage optimization of engineering resources based on access popularity includes the following steps:

[0008] S1. Acquire project site data, identify the current construction stage based on the site data, where the construction stage includes the main structure stage, the secondary structure stage, and the decoration installation stage, and calculate the dynamic popularity of each component, i.e., the access popularity;

[0009] S2. Build a distributed resource scheduling network, connect all scheduling nodes through the Internet of Things, synchronize data in real time, and transmit it;

[0010] S3. Obtain map information of the construction site and configure the warehouse partition topology based on the current construction stage and project resource access popularity to obtain the construction site partition results.

[0011] S4 receives data from all scheduling nodes in S2 and the partitioning results of the construction site in S3, and dynamically selects the scheduling mode for engineering resources based on the current construction phase type and engineering resource access popularity;

[0012] S5. Based on the engineering site data from S1, monitor whether there are any abnormal construction events. Based on whether there are any abnormal events, determine whether to trigger access heat reassessment and schedule storage optimization.

[0013] Preferably, the specific working steps of S1 are as follows:

[0014] S11. Acquire project site data, specifically, obtain the project BIM schedule, UWB positioning data, and construction machinery current signals;

[0015] The acquired project BIM schedule, UWB positioning data, and construction machinery current signals are pre-processed as project site data;

[0016] S12, performing stage feature extraction, inputting the pre-processed BIM schedule, UWB positioning data, and construction machinery current signal;

[0017] Output stage label , the stage label is used to identify the current construction stage;

[0018] S13. Weight matrix of component stage:

[0019] Create a three-dimensional weight tensor ;

[0020] Where p is the stage label, i is the component category, They represent the urgency weight, frequency weight, difficulty weight and importance weight of prefabricated component category i under stage label p respectively;

[0021] S14. For each prefabricated component j, obtain the standardized urgency vector, frequency vector, difficulty vector, and importance vector. According to its stage label p and component category i, calculate its dynamic heat according to the weight matrix and the standardized urgency vector, frequency vector, difficulty vector, and importance vector. ;

[0022] S15, monitoring the changes of the stage label p in real time, and triggering a stage transition response when detecting that the current stage label p is different from the construction stage at the previous moment;

[0023] Reset dynamic heat for all precast components , recalculate the weight matrix according to the new construction stage , and synchronize network data to update the latest construction stage information and component dynamics The information is communicated to all scheduling nodes.

[0024] Preferably, the specific working steps of S2 are as follows:

[0025] Through the Internet of Things technology, prefabricated component manufacturers, transportation vehicles and warehouse nodes are interconnected to establish a real-time data channel;

[0026] The manufacturer node is responsible for providing production progress data, inventory data and quality inspection data of prefabricated components; the transport vehicle node is responsible for providing the vehicle's location, speed, type and quantity of loaded components, and estimated arrival time; the warehouse node is responsible for providing the warehouse's inventory data, storage location information, warehouse environment data, including temperature and humidity, and inbound and outbound record data.

[0027] Preferably, the specific working steps of S3 are as follows:

[0028] S31. Obtain the dynamic heat of the current stage label p and each prefabricated component j, and collect the physical properties of the construction site map, including the foundation load and slope of the site;

[0029] S32, based on the current stage label p and the dynamic heat of each prefabricated component j , calculate the access heat field , represents the access heat distribution of each prefabricated component j at different locations in the construction site;

[0030] Then, based on the physical properties of the construction site map, the stress field is obtained. , which indicates the restrictions imposed on warehouse partitions by the physical conditions of the construction site;

[0031] Will visit the heat field and stress field The coupled field is obtained by linear combination ;

[0032] S33, in the coupled field Peak point deployment partition seeds, when the coupling field When the value of is greater than the preset threshold, a partition seed is created in the partition;

[0033] The radius of the seed point is determined according to the maximum access popularity of the component;

[0034] Get the construction site plan map, divide the construction site plan map into N×Y grids, initialize the state of each grid to undefined, for each map grid in the construction site , calculate its attribution tendency value t;

[0035] Get the threshold of t, compare it with the threshold according to the value of t, mark the partitions according to the comparison results, obtain the partition results of each map grid, map them on the construction site plane map, and obtain the final grid map state;

[0036] Convert the final raster map state into vector boundaries, extract the boundaries of different partitions, represent different partitions with different colors on the construction site plan map, and mark the attributes of each partition on the construction site plan map.

[0037] Preferably, the S33 further includes the following steps:

[0038] When the stage label p is switched, the size of each partition on the construction site plan map is adjusted according to the maximum component access heat ratio between the current stage and the previous stage.

[0039] Preferably, the specific working steps of S4 are as follows:

[0040] S41. Based on the access heat field and stress field Calculate the partition heat gradient difference and heat change rate;

[0041] S42. Obtain a reference dimension, and automatically select a scheduling mode based on the comparison of the partition heat gradient difference multiplied by the heat change rate. If the partition heat gradient difference multiplied by the heat change rate is greater than the reference dimension, select the first scheduling mode; otherwise, select the second scheduling mode.

[0042] Preferably, the S4 further includes S43, which is used to drive the scheduling execution according to the selected scheduling mode, and the specific steps are as follows:

[0043] If the first scheduling mode is used, the product of the visit popularity and the inventory ratio at each manufacturer is calculated, and the manufacturer with the largest product is selected. The visit popularity is the visit popularity at the manufacturer's location, and the inventory ratio is the manufacturer's inventory sufficiency, that is, the percentage of the current inventory divided by the maximum inventory;

[0044] When the vehicle reaches the target partition and the time derivative of the access heat gradient is zero, unloading begins;

[0045] If it is the second scheduling mode, first calculate the transportation frequency, during each transportation process;

[0046] Find the point with the smallest access heat gradient in the heavy area as the unloading point;

[0047] Calculate the partial derivative of the access heat in the height direction z at each component position in the vertical component area, and sort the components in descending order according to this value.

[0048] Preferably, the S5 comprises the following steps:

[0049] S51. In the access heat field In , the ratio of planned access heat to actual access heat is calculated at each location, and the natural logarithm is taken and weighted summed to obtain an overall difference measure;

[0050] Obtain system parameters. When the overall difference metric is greater than the system parameters, it is determined that the project progress is lagging behind, and an alarm is triggered.

[0051] Obtain the activity of the manufacturer's node. If the activity of the manufacturer's node is lower than the first quartile of the historical activity data, it is determined that the manufacturer's node has been interrupted, and an alarm is triggered;

[0052] The stress field mutation rate is calculated by calculating the rate of change of the stress field gradient over time. If the stress field mutation rate exceeds the historical average plus two times the standard deviation, an alarm is triggered.

[0053] Preferably, the S5 further includes S52, which is used to re-evaluate the access popularity after receiving the alarm. The specific steps are as follows:

[0054] After receiving the alarm, if the project progress is lagging behind, the updated heat value is calculated based on the overall difference measurement ;

[0055] If it is the other two reasons of S51, the updated heat value is calculated based on the stress field mutation rate .

[0056] Preferably, the specific steps of S52 further include the following:

[0057] The updated heat value Enter S3 to recalculate the access popularity field and trigger the reselection of the scheduling mode;

[0058] Call S4 to reselect the scheduling mode and drive the scheduling execution according to the selected scheduling mode.

[0059] The present invention provides a method for distributed scheduling and storage optimization of engineering resources based on access popularity. It has the following beneficial effects:

[0060] 1. By acquiring project site data to identify construction stages and calculate the dynamic popularity of components, the distributed component resource scheduling network achieves real-time synchronous data transmission. The scheduling mode is dynamically selected based on the construction stage type and the access popularity of project resources. This allows for precise resource allocation based on actual construction needs, avoiding idle resources and waste, and improving the utilization efficiency of project resources.

[0061] 2. Real-time monitoring of changes in the construction phase and abnormal events, triggering access popularity re-evaluation and scheduling storage optimization, can timely adjust resource allocation and scheduling strategies when abnormal situations such as project progress delays, manufacturer node interruptions or stress field mutations occur, ensuring the project progresses on time, reducing the adverse effects of abnormalities on construction, and improving the stability and reliability of the construction process. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0063] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0064] like Figure 1 The present invention proposes a method for distributed scheduling and storage optimization of engineering resources based on access popularity, comprising the following steps:

[0065] S1. Acquire project site data, identify the current construction stage based on the site data, where the construction stage includes the main structure stage, the secondary structure stage, and the decoration installation stage, and calculate the dynamic popularity of each component, i.e., the access popularity;

[0066] The implementation of this method is highly dependent on the collection, processing and integration of multi-source heterogeneous data throughout the entire project life cycle. The specific data categories to be integrated include but are not limited to:

[0067] Project basic data and location data: including project site coordinates, site topographic map, building structure information and other basic spatial and attribute data;

[0068] Construction plans and schemes: The core is the structured engineering BIM schedule, while also considering integrating some unstructured construction plan documents (such as special construction plans and technical briefing records) to provide a basis for phase identification and resource demand forecasting. For unstructured data (such as non-geometric information in the BIM model and scheme text), key elements (such as key nodes, resource demand descriptions, and process requirements) are extracted through natural language processing (NLP) and information extraction technology.

[0069] Material data: inventory status (quantity, location), inbound and outbound records of prefabricated components and other building materials, in-transit transportation information (from the S2 dispatch network), quality inspection reports, etc.

[0070] Equipment data: real-time status of construction machinery (such as current signal, GPS / UWB location, working hours), equipment records, maintenance records, etc.

[0071] On-site perception data: including UWB high-precision positioning data, environmental sensor data (temperature and humidity), and auxiliary on-site monitoring video data (can be used for visual verification, security monitoring, and indirectly assisting in activity recognition);

[0072] Project quality and safety data and specifications: quality inspection records, acceptance results, safety hazard reports, rectification notices, relevant construction specifications and standard requirements;

[0073] The above multi-source data is aggregated and synchronized in real time through the distributed resource scheduling network of the S2 component. Structured data (such as BIM progress, inventory, positioning coordinates, and equipment signals) is directly used for calculations (such as stage identification, dynamic heat, and access heat field). Unstructured or semi-structured data (such as partial BIM information, construction plans, and quality reports) are converted into structured features or weight factors after preprocessing and information extraction. They are input into the stage weight matrix component, anomaly monitoring (S51), or used as constraints to influence warehouse partitioning (stress field) and scheduling decisions. The master data serves as the core dictionary of the entire system, ensuring the consistency of resource identification and the accuracy of scheduling. The integration of all data provides comprehensive and real-time information support for dynamic scheduling and storage optimization based on access heat.

[0074] S2. Build a distributed resource scheduling network that connects all scheduling nodes through the Internet of Things, synchronizes data in real time, and transmits it. It should be noted that scheduling nodes specifically include manufacturer locations, transport vehicles, and construction site warehouse capacity.

[0075] S3. Obtain map information of the construction site and configure a warehouse partition topology based on the current construction stage and access popularity of engineering resources to obtain a partition result for the construction site; the warehouse partition topology includes a heavy area, a vertical storage area, and an indoor storage area;

[0076] S4 receives data from all scheduling nodes in S2 and the partitioning results of the construction site in S3, and dynamically selects the scheduling mode for engineering resources based on the current construction phase type and engineering resource access popularity;

[0077] S5. Based on the engineering site data from S1, monitor whether there are any abnormal construction events. Based on whether there are any abnormal events, determine whether to trigger access heat reassessment and schedule storage optimization.

[0078] As an optional embodiment: the specific working steps of S1 are as follows:

[0079] S11. Acquire project site data, specifically, obtain the project BIM schedule, UWB positioning data, and construction machinery current signals. It should be noted that the specific acquisition method is as follows:

[0080] The BIM schedule is usually prepared by the project management team in the early stages of the project and stored in the project's BIM management software or project management information system. Through the data interface with these systems, such as using the API (Application Programming Interface) or directly exporting files in MSProject format, it is imported into the data analysis system used in this method to obtain BIM schedule data.

[0081] UWB positioning base stations will be deployed at the construction site. These base stations are fixedly installed on the buildings or scaffolding at the construction site according to a certain spatial layout. Target objects such as personnel, equipment, and prefabricated components wear or install UWB positioning tags. The UWB positioning base stations communicate with the positioning tags through wireless signals and obtain the location information of the tags in real time.

[0082] Install current transformers and other current signal sensors in the power circuit of the construction machinery. These sensors can sense the current changes of the construction machinery in real time and obtain the current signal of the construction machinery at a set sampling rate (1kHz).

[0083] The acquired engineering BIM schedule, UWB positioning data, and construction machinery current signals are preprocessed as engineering site data. It should be noted that for the BIM schedule data, the task schedule is parsed, and the tasks related to the construction of prefabricated components and their time stamps are extracted. The UWB positioning data is subjected to coordinate correction and data filtering to remove positioning errors and abnormal positioning points, and the effective position trajectories of construction personnel, equipment, and prefabricated components are extracted. The construction machinery current signals are subjected to signal filtering and feature extraction to remove noise interference in the signals and extract characteristic parameters that can reflect the operating status of the construction machinery, such as the effective value, root mean square value, and peak value of the current.

[0084] S12, performing stage feature extraction, inputting the pre-processed BIM schedule, UWB positioning data, and construction machinery current signal;

[0085] Output stage label The stage label is used to identify the current construction stage. It should be noted that the BIM schedule provides a detailed schedule for the entire project, including information such as the start time, end time, and duration of each construction task. On-site UWB positioning data uses UWB positioning devices installed on site to obtain real-time location information of construction personnel, equipment, and prefabricated components. The current signal of construction machinery reflects the operating status and workload of the construction machinery.

[0086] First, the BIM schedule is parsed to extract construction tasks related to prefabricated components and their schedules. The system then combines on-site UWB positioning data with construction machinery current signals to analyze the characteristics and progress of current construction activities. The UWB positioning data is used to determine the location distribution of construction personnel and equipment on the construction site, and to determine the current construction area and stage. The current signals from construction machinery are then used to determine whether the working status of the machinery meets the requirements of the current construction stage.

[0087] Based on the above analysis, determine the label of the current construction stage;

[0088] For example, when the construction site is mainly engaged in the installation of load-bearing components such as columns and beams, and the construction machinery is mainly large-scale lifting equipment, it is judged to be in the main structure stage;

[0089] S13. Weight matrix of component stage:

[0090] Create a three-dimensional weight tensor ;

[0091] Where p is the stage label, i is the component category, They represent the urgency weight, frequency weight, difficulty weight and importance weight of prefabricated component category i under stage label p respectively;

[0092] It should be noted that prefabricated components specifically include concrete, steel structures and non-load-bearing components;

[0093] It should also be noted that the specific weights are obtained through a neural network model, which collects historical construction data, including the actual values ​​of urgency, frequency, difficulty, and importance of different construction stages and component categories, as well as the corresponding result data such as construction progress, resource allocation, and cost.

[0094] Clean and normalize the data to obtain the input feature matrix X and target vector Y;

[0095] Create a neural network model with 4 neurons in the input layer (corresponding to urgency, frequency, difficulty, and importance), several neurons in the hidden layer, and 4 neurons in the output layer (corresponding to weight coefficients).

[0096] Use the training set to train the model;

[0097] Input new construction data into the trained neural network model, and the model will automatically calculate the relative weight of each weight factor based on the input data;

[0098] For example, input the urgency, frequency, difficulty and importance values ​​of a new main structure period concrete component, and the model outputs its weight vector;

[0099] S14. For each prefabricated component j, obtain the standardized urgency vector, frequency vector, difficulty vector, and importance vector. According to its stage label p and component category i, calculate its dynamic heat according to the weight matrix and the standardized urgency vector, frequency vector, difficulty vector, and importance vector. ;

[0100] It should be noted that, in this embodiment, the specific calculation formula may be:

[0101] ;

[0102] in are the standardized urgency vector, frequency vector, difficulty vector, and importance vector;

[0103] The specific way to obtain it is:

[0104] Obtain the urgency value of each component from the construction plan and actual progress data, specifically the ratio of remaining time to planned time;

[0105] Obtain the frequency value of each component from IoT sensors or construction management systems, specifically the number of times it is used per day;

[0106] The daily usage count of each component is obtained through UWB positioning tags for large components and assembly components, and small components and other large components are obtained based on the consumption table of the monthly component procurement plan.

[0107] Obtain the construction difficulty of each component from the construction process document. Calculate the surface area to volume ratio of the component using the built-in functions of the BIM software or by writing a simple script. The formula is: surface area to volume ratio. The higher the ratio, the more complex the component shape and the greater the construction difficulty. This ratio is used as the difficulty value.

[0108] The BIM model is integrated with the structural analysis software. The stress conditions of the components are calculated through the structural analysis software. Based on indicators such as the force magnitude and stress distribution, the components that have a significant impact on the structural stability are determined. The importance value of components such as columns and beams that bear the main loads is set to 1.0, and the initial value, that is, the non-load-bearing components, is 0.5.

[0109] Calculate the minimum and maximum values ​​of each indicator (urgency, frequency, difficulty, importance), perform normalization, and use the Min-Max normalization method to map the value of each indicator to the [0,1] interval;

[0110] For each component j, its normalized urgency, frequency, difficulty, and importance values ​​are combined into a vector, namely, vector ;

[0111] S15, monitoring the changes of the stage label p in real time, and triggering a stage transition response when detecting that the current stage label p is different from the construction stage at the previous moment;

[0112] Reset dynamic heat for all precast components , recalculate the weight matrix according to the new construction stage , and synchronize network data to update the latest construction stage information and component dynamics The information is communicated to all scheduling nodes so that each node can adjust resource scheduling and storage strategies in a timely manner;

[0113] It should be noted that access popularity refers to the frequency and intensity of access by construction personnel, equipment or systems to information or resources related to specific prefabricated components during the construction process, reflecting the degree of attention and demand for the component in the current construction stage.

[0114] As an optional embodiment: the specific working steps of S2 are as follows:

[0115] Through the Internet of Things technology, prefabricated component manufacturers, transportation vehicles and warehouse nodes are interconnected to establish a real-time data channel;

[0116] The manufacturer node is responsible for providing production progress data, inventory data and quality inspection data of prefabricated components; the transport vehicle node is responsible for providing the vehicle's location, speed, type and quantity of loaded components, and estimated arrival time; the warehouse node is responsible for providing the warehouse's inventory data, storage location information, warehouse environment data, including temperature and humidity, and inbound and outbound record data.

[0117] It is also necessary to formulate a unified data communication protocol to ensure that data transmission between nodes is accurate, timely and reliable. The data communication protocol specifically includes data format, data transmission frequency, data encryption method and data verification method to meet the data communication needs in the complex environment of the engineering site.

[0118] As an optional embodiment: the specific working steps of S3 are as follows:

[0119] S31. Obtain the dynamic heat of the current stage label p and each prefabricated component j, and collect the physical properties of the construction site map, specifically including the site's foundation load-bearing capacity and slope. It should be noted that foundation load-bearing capacity is usually expressed in units of pressure. The International System of Units (SI) unit is Pascal (Pa), where 1Pa = 1N / m². Specifically, soil samples obtained on-site are brought to the laboratory for physical and mechanical property testing. For example, a triaxial compression test applies stress in different directions to the soil sample, measuring mechanical indicators such as shear strength. The bearing capacity of the foundation is then calculated based on theoretical formulas for soil mechanics.

[0120] Slope can be measured using a level. First, set up several measuring points on the site, install a ruler on the level, and calculate the slope by measuring the height difference between different points. The unit of slope is usually expressed as a percentage.

[0121] S32, based on the current stage label p and the dynamic heat of each prefabricated component j , calculate the access heat field , represents the access heat distribution of each prefabricated component j at different locations in the construction site; it should be noted that according to the plan of the construction site, a plane coordinate system is established to obtain the position coordinates of each point in the site , for each prefabricated component, determine its installation position, and then obtain its installation position coordinates ;

[0122] The specific calculation formula for access heat field is: ,in is the thermal attenuation coefficient, which is specifically set to 50m in this embodiment and can be adjusted as needed;

[0123] This formula represents the access popularity of the component As the distance from the installation location increases, the attenuation is determined by By this formula, we can calculate the location of each position in the field. The thermal value of the area with high thermal value indicates that the precast components are visited more frequently;

[0124] This method can construct an access heat field model to quantify the heat of different locations in the site, facilitating subsequent planning of warehouse partitions, etc.

[0125] Then, based on the physical properties of the construction site map, the stress field is obtained. , which represents the restrictions imposed by the physical conditions of the construction site on the warehouse partitions; it should be noted that, in this embodiment, the calculation formula for the stress field is ;

[0126] Here It is the Laplace operator, which is used to calculate the second-order derivatives of load bearing and slope, thereby obtaining the distribution of stress field, which reflects the influence of the site's bearing capacity and terrain slope on the warehouse partition;

[0127] Where 0.3 is the weight coefficient, which is the default initial value in this embodiment. Specifically, the weight coefficient can be changed to observe the changes in stress field distribution and the impact on the warehouse partition results, and the weight value that makes the partition results most reasonable can be selected;

[0128] Will visit the heat field and stress field The coupled field is obtained by linear combination ; It should be noted that the coupling field calculation formula is ;

[0129] in is the stage weight. In this embodiment, the main structure stage takes 1.5, and the other stages take 0.8. Is a balance coefficient used to adjust the relative influence of the access heat field and stress field on the total coupled field. , the weights of access heat field and stress field in partition decision can be controlled;

[0130] The value of can be determined by the optimization model, and the optimal The value is obtained by regression analysis to ensure that the coupling field can best reflect the rationality and stability of the warehouse partition;

[0131] S33, in the coupled field Peak point deployment partition seeds, when the coupling field When the value of is greater than the preset threshold, a partition seed is created in the partition;

[0132] The radius of the seed point is determined according to the maximum access popularity of the component;

[0133] Get the construction site plan map, divide the construction site plan map into N×Y grids, initialize the state of each grid to undefined, for each map grid in the construction site , calculate its attribution tendency value t; it should be noted that the specific calculation formula is:

[0134] ;

[0135] Get the threshold of t, compare it with the threshold according to the value of t, mark the partition according to the comparison result, get the partition result of each map grid, map it on the plane map of the construction site, and get the final grid map state; It should be noted that the threshold of t is specifically divided into the first threshold, the second threshold and the third threshold. If t is greater than the first threshold, it is marked as a heavy area. If t is less than or equal to the first threshold and greater than the second threshold, it is marked as a vertical storage area. If the stress field is If it is greater than the third threshold, it is marked as an indoor storage area;

[0136] The specific calculation steps of the first threshold are:

[0137] The priority value is determined according to the current construction stage. The priority of the main structure period is 3, the priority of the secondary structure period is 1, and the priority of the decoration and installation period is 0.5. The access heat of all components is averaged to obtain the average access heat. The dispersion of the access heat is measured by the following calculation formula: the square root of the average of the square of the difference between the heat value and the average value. The base value is fixed at 0.5. The stage priority is divided by 3 and then multiplied by 0.2 to obtain the stage weight part. First, the difference between the maximum access heat and the average access heat is calculated, and then divided by the heat standard deviation. The result is substituted into the hyperbolic tangent function and finally multiplied by 0.3 to obtain the heat deviation part.

[0138] Add the base value, the stage weight part, and the heat deviation part to obtain the first threshold;

[0139] The demand for heavy-duty zones varies across construction phases. The main structure phase requires more zones to store heavy components, while the decorative installation phase requires less. The trigger threshold for the heavy-duty zone is dynamically adjusted through stage weights to meet the construction needs of different phases. Considering the discrete degree of component access popularity, components with high access popularity are prioritized for storage in the heavy-duty zone to improve construction efficiency. The hyperbolic tangent function ensures that the heat dispersion varies within a reasonable range.

[0140] The calculation steps of the second threshold are:

[0141] Calculate the 70th percentile of the site's load-bearing map to reflect the statistical value of the site's load-bearing capacity. Determine the effective area with a slope less than 5°. Calculate the total volume of precast beams and columns at the current stage. First, divide the total volume of heavy components by the effective area, multiply by 0.5, and finally add 1. Multiply the 70th percentile of the site's load-bearing map by the adjustment coefficient to obtain a preliminary load-bearing requirement. Take the smaller of the preliminary load-bearing requirement and 15 to obtain the final second threshold.

[0142] The base value of the third threshold is 0.3, which can be adjusted as needed in actual use:

[0143] In other cases, keep the map grid undefined;

[0144] The final raster map state is converted into vector boundaries, and the boundaries of different zones are extracted. Different zones are represented by different colors on the construction site plan map, and the attributes of each zone are annotated on the construction site plan map. It should be noted that in this embodiment, the different zones include heavy-duty areas, vertical storage areas, and indoor storage areas.

[0145] As an optional embodiment: S33 further includes the following steps:

[0146] When the stage label p is switched, the size of each partition on the construction site plan map is adjusted according to the maximum component access heat ratio between the current stage and the previous stage.

[0147] It should be noted that, in this embodiment, the specific adjustment steps are:

[0148] Get the initial area of ​​each partition area ;

[0149] The number of partitions is 3, and the value of s is also 1 to 3;

[0150] For heavy duty areas:

[0151] During the main structure period, the demand for heavy components is extremely high and the access popularity is high, so the scope of the heavy area should be large to meet the storage needs of a large number of heavy components.

[0152] During the secondary structure period: compared with the main structure period, the demand for heavy components is reduced and the access heat is reduced, so the scope of the heavy area should be reduced to free up space for other partitions;

[0153] During the decoration installation period: the demand for heavy components will further decrease, the popularity of visits will continue to decrease, and the scope of the heavy area should be further reduced;

[0154] It should also be noted that the specific calculation formula is based on the initial area of ​​the heavy zone , the calculation formula is:

[0155] For vertical storage areas, it is used to store light components;

[0156] During the main structure stage: the demand for lightweight components is relatively small, and the scope of the vertical storage area can be smaller;

[0157] During the secondary structure period: the demand for lightweight components such as partition boards increases, the popularity of visits increases, and the scope of vertical storage areas should be expanded;

[0158] During the decoration and installation period: the demand for lightweight components remains at a high level, and the scope of vertical storage areas should remain large or continue to expand;

[0159] For indoor areas;

[0160] During the main structure phase: there is relatively little demand for environmentally sensitive components, and the indoor area can be smaller;

[0161] During the secondary structure period: the demand for environmentally sensitive components increases, the popularity of visits rises, and the scope of indoor areas should be expanded;

[0162] During the decoration and installation period: the demand for environmentally sensitive components will further increase, the popularity of visits will remain at a high level, and the scope of the indoor area should be further expanded;

[0163] ;

[0164] in Indicates the area of ​​the reload zone after adjustment, is the initial area, that is , is the maximum component access heat in the previous stage, The maximum component access heat in the current stage;

[0165] xs is the adjustment coefficient, and the value of s ranges from 1 to 3, one to one correspondence , the specific value is:

[0166] Collect visitor popularity data and corresponding zoning adjustments for similar projects in the past at different construction stages, including visitor popularity and zoning size changes for heavy, light, and indoor areas;

[0167] For each zone, calculate the historical adjustment ratio when different construction phases are switched. For example, for the heavy zone, calculate the change ratio of its visit heat from the main structure period to the secondary structure period.

[0168] According to these change ratios, the adjustment coefficient of each partition when it transitions to different stages is determined. For example, if the visit popularity of the heavy zone drops by an average of 30% from the main structure period to the secondary structure period, the adjustment coefficient can be set to 0.7.

[0169] For the three partitions, the adjustment coefficients are different;

[0170] It should be noted that through this dynamic adjustment, it can be ensured that the warehouse partitions always match the construction progress and needs, thereby improving the utilization efficiency of storage space and the convenience of material management. Timely reduction of the scope of heavy areas can free up space for other types of partitions, reduce the secondary handling of materials, and improve construction efficiency.

[0171] As an optional embodiment: the specific working steps of S4 are as follows:

[0172] S41. Based on the access heat field and stress field Calculate the partition heat gradient difference and heat change rate;

[0173] It should be noted that the specific calculation steps for the partition heat gradient difference and heat change rate are as follows:

[0174] For the zone heat gradient difference, the absolute value of the difference between the maximum gradient of the heavy zone and the vertical zone is calculated to reflect the difference in the intensity of the access heat change between the two areas. It is used to measure the contrast of heat changes between different zones and provide a basis for mode switching;

[0175] For the heat change rate, according to the formula , calculate the heat change rate;

[0176] Where T is the duration of the current construction phase, which comes from the time planning data of the S1 phase. To calculate the partial derivative of the access heat field over time, which represents the rate of change of access heat over time;

[0177] S42. Obtain a reference dimension, and automatically select a scheduling mode based on the comparison of the partition heat gradient difference multiplied by the heat change rate. If the partition heat gradient difference multiplied by the heat change rate is greater than the reference dimension, select the first scheduling mode; otherwise, select the second scheduling mode.

[0178] It should be noted that the first scheduling mode is the emergency scheduling mode, and the second scheduling mode is the circular supply mode, which is used for normalized logistics scheduling;

[0179] It should also be noted that the specific calculation formula for the benchmark dimension is: ;

[0180] in Combined with the Boltzmann constant and reduced Planck constant , forming a quantum thermodynamic characteristic frequency, which is used to combine the thermodynamic and quantum properties of the system to form a reference dimension;

[0181] The natural logarithm of the ratio of the number of manufacturers, the number of vehicles, and the number of warehouses is calculated to comprehensively reflect the quantitative relationship of resources in the logistics system. This ratio is multiplied by the aforementioned quantum thermodynamic characteristic frequency to obtain the critical value of mode switching.

[0182] The partition heat gradient difference reflects the spatial imbalance of the system, and the heat change rate reflects the temporal instability of the system. The product constitutes the spatiotemporal instability index. When the spatiotemporal instability exceeds the quantum fluctuation threshold, the system requires strong intervention. When the spatiotemporal instability is within the range of natural fluctuations, the system can be self-sustaining.

[0183] As an optional embodiment, the step S4 further includes step S43, which is used to drive the scheduling execution according to the selected scheduling mode. The specific steps are as follows:

[0184] If the first scheduling mode is used, the product of the visit popularity and the inventory rate at each manufacturer is calculated, and the manufacturer with the largest product is selected. The visit popularity refers to the visit popularity at the manufacturer's location, and the inventory rate refers to the manufacturer's inventory sufficiency, that is, the percentage of the current inventory divided by the maximum inventory. It should be noted that the visit popularity is calculated by counting the number of times each manufacturer is visited within a certain time range. In this embodiment, the number of times each manufacturer is visited by logistics vehicles in the past month is recorded.

[0185] When the vehicle reaches the target zone and the time derivative of the access heat gradient reaches zero, unloading begins. It should be noted that a zero time derivative of the access heat gradient indicates that the access heat at that location has reached a relatively stable state. Unloading at this time can avoid operations in areas with drastic changes in access heat, reducing potential confusion and efficiency losses.

[0186] If it is the second scheduling mode, first calculate the transportation frequency during each transportation process; it should be noted that the calculation formula steps of the transportation frequency are:

[0187] Set a reference angular frequency by collecting the system's transportation frequency data over a period of time, calculating its average or median, calculating the average Laplace value of the access heat field, and performing Laplace operator calculation on the access heat field to obtain the Laplace value of each position;

[0188] The Laplace values ​​at all locations are averaged to obtain the average Laplace value of the access heat field, which is recorded as the average gradient. The reference Laplace value is determined, which is recorded as the reference gradient. Specifically, the distribution of the Laplace values ​​of the access heat field during the stable operation of the system is analyzed, and the value corresponding to the peak of the distribution is selected as the reference gradient.

[0189] Divide the average gradient by the reference gradient to obtain the normalized gradient ratio;

[0190] Take the square root of the gradient ratio, divide the reference angular frequency by 2π to get the reference frequency, and multiply the reference frequency by the square root to get the final transport frequency;

[0191] Find the point with the smallest visit heat gradient in the heavy-duty area as the unloading point. A location with a small visit heat gradient means that the visit heat changes smoothly and the logistics activities are relatively stable, which is suitable as an unloading point, helping to improve unloading efficiency and reduce possible confusion.

[0192] Calculate the partial derivative of access popularity at each component location within the vertical component area with respect to height z, and sort the components in descending order based on this value. This partial derivative reflects the rate of change of access popularity with height. By sorting in descending order, we can better adapt to the spatial layout characteristics of the vertical area, improve the efficiency of material storage and retrieval, and facilitate subsequent transportation and use.

[0193] As an optional embodiment: S5 includes the following steps:

[0194] S51. In the access heat field In , the ratio of planned access heat to actual access heat is calculated at each location, and the natural logarithm is taken and weighted summed to obtain an overall difference measure;

[0195] Obtain system parameters. When the overall difference metric is greater than the system parameters, it is determined that the project progress is lagging behind, and an alarm is triggered.

[0196] It should be noted that the system parameter value is ;

[0197] Obtain the activity of the manufacturer node. If the activity of the manufacturer node is lower than the first quartile of the historical activity data, it is determined that the manufacturer node has been interrupted and an alarm is triggered. It should be noted that the calculation formula for the manufacturer node activity is ; Where u is the plane position coordinate of the manufacturer;

[0198] The stress field mutation rate is calculated by calculating the rate of change of the stress field gradient over time. If the stress field mutation rate exceeds the historical average plus two standard deviations, an alarm is triggered. It should be noted that by calculating the ratio of planned to actual access popularity and measuring the difference, combined with monitoring of manufacturer node activity and stress field mutation rate, multi-dimensional control of project progress and on-site conditions is achieved. When various indicators exceed the normal range, an alarm is triggered, allowing construction management personnel to respond quickly and take measures to prevent further delays in progress or supply chain disruptions from having a greater impact on the project.

[0199] Compared with the existing technology that relies solely on progress plans or simply monitors some indicators, it can more accurately reflect the actual progress of the project and avoid monitoring blind spots caused by focusing only on local areas.

[0200] As an optional embodiment, the step S5 further includes step S52, which is configured to re-evaluate access popularity after receiving an alarm. The specific steps are as follows:

[0201] After receiving the alarm, if the project progress is lagging behind, the updated heat value is calculated based on the overall difference measurement It should be noted that the specific calculation steps are to calculate the average criticality, read the criticality flag of all components (1 indicates a critical component, 0 indicates a non-critical component), accumulate the criticality values ​​of all components, and divide it by the total number of components;

[0202] Take the original value , calculate the difference ratio: (difference metric - system parameter) / system parameter;

[0203] Calculate the exponential function: e to the power of (difference ratio);

[0204] Calculate the criticality factor: 1 + (current component criticality / average criticality);

[0205] The original value × exponential function × critical factor to get the updated heat value ;

[0206] If it is the other two reasons of S51, the updated heat value is calculated based on the stress field mutation rate It should be noted that the specific calculation steps are to determine the intrinsic mutation rate, calculate the average mass density of the site, calculate the total mass of all components, divide it by the total area of ​​the site, take the gravitational constant G, calculate G×average density, divide it by the square of the speed of light c, and take the square root of the result;

[0207] Then list the location coordinates of all partitioned warehouses, calculate the straight-line distance between each pair of warehouses, accumulate the distances of all warehouse pairs, and divide by the total number of warehouse pairs to obtain the system scale;

[0208] Take the stress field mutation rate, divide it by the intrinsic mutation rate, measure the distance from the component location to the center of the abnormal event (if it is a manufacturer, it is the distance from the manufacturer to the warehouse; if it is an environmental anomaly, it is the distance from the center of the abnormal environment to the warehouse) and divide it by the system scale. Multiply the two ratios to obtain the normalized impact factor.

[0209] Take the original value , calculate the negative normalized impact factor, calculate the power of (negative normalized impact factor) of e, and convert the original value Multiply by e to the power of (negative normalized impact factor) to get the updated heat value ;

[0210] Depending on the cause of the alarm, the overall difference measurement or the stress field mutation rate and distance are used to calculate the updated heat value, making the adjustment of access heat more targeted and better reflecting the changes in actual project needs. After receiving the alarm, the access heat is re-evaluated in a timely manner, providing an accurate basis for subsequent scheduling adjustments and enhancing the system's adaptability to dynamic changes.

[0211] As an optional embodiment: the specific step of S52 further includes the following:

[0212] The updated heat value Enter S3 to recalculate the access popularity field and trigger the reselection of the scheduling mode;

[0213] S4 is called to reselect the scheduling mode and drive the execution of the selected scheduling mode. The updated heat value is input into S3 to recalculate the access heat field, trigger the reselection of the scheduling mode and the execution of the scheduling. This forms a complete closed-loop optimization process, ensuring the continuous optimization of resource scheduling and storage strategies. The automated reselection of the scheduling mode and the driving of the scheduling execution reduce manual intervention, improve the efficiency of decision-making and execution, and help accelerate the project progress.

[0214] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for distributed scheduling and storage optimization of engineering resources based on access popularity, characterized in that: The following steps are involved: S1. Acquire project site data, identify the current construction stage based on the site data, where the construction stage includes the main structure stage, the secondary structure stage, and the decoration installation stage, and calculate the dynamic popularity of each component, i.e., the access popularity; S2. Build a distributed resource scheduling network, connect all scheduling nodes through the Internet of Things, synchronize data in real time, and transmit it; S3. Obtain map information of the construction site and configure the warehouse partition topology based on the current construction stage and project resource access popularity to obtain the construction site partition results. S4 receives data from all scheduling nodes in S2 and the partitioning results of the construction site in S3, and dynamically selects the scheduling mode for engineering resources based on the current construction phase type and engineering resource access popularity; S5, based on the engineering site data from S1, monitors whether there are any abnormal construction events. Based on whether there are any abnormal events, it determines whether to trigger access popularity re-evaluation and scheduling storage optimization; The specific working steps of S1 are as follows: S11. Acquire project site data, specifically, obtain the project BIM schedule, UWB positioning data, and construction machinery current signals; The acquired project BIM schedule, UWB positioning data, and construction machinery current signals are pre-processed as project site data; S12, performing stage feature extraction, inputting the pre-processed BIM schedule, UWB positioning data, and construction machinery current signal; Output stage label , the stage label is used to identify the current construction stage; S13. Weight matrix of component stage: Create a three-dimensional weight tensor ; Where p is the stage label, i is the component category, They represent the urgency weight, frequency weight, difficulty weight and importance weight of prefabricated component category i under stage label p respectively; S14. For each prefabricated component j, obtain the standardized urgency vector, frequency vector, difficulty vector, and importance vector. According to its stage label p and component category i, calculate its dynamic heat according to the weight matrix and the standardized urgency vector, frequency vector, difficulty vector, and importance vector. ; S15, monitoring the changes of the stage label p in real time, and triggering a stage transition response when detecting that the current stage label p is different from the construction stage at the previous moment; Reset dynamic heat for all precast components , recalculate the weight matrix according to the new construction stage and synchronized with network data to update the latest construction phase information and component dynamics The information is communicated to all scheduling nodes; The specific working steps of S3 are as follows: S31. Obtain the dynamic heat of the current stage label p and each prefabricated component j, and collect the physical properties of the construction site map, including the foundation load and slope of the site; S32, based on the current stage label p and the dynamic heat of each prefabricated component j , calculate the access heat field , represents the access heat distribution of each prefabricated component j at different locations in the construction site; Then, based on the physical properties of the construction site map, the stress field is obtained. , which indicates the restrictions imposed on warehouse partitions by the physical conditions of the construction site; Will visit the heat field and stress field The coupled field is obtained by linear combination ; S33, in the coupled field Peak point deployment partition seeds, when the coupling field When the value of is greater than the preset threshold, a partition seed is created in the partition; The radius of the seed point is determined according to the maximum access popularity of the component; Get the construction site plan map, divide the construction site plan map into N×Y grids, initialize the state of each grid to undefined, for each map grid in the construction site , calculate its attribution tendency value t; Get the threshold of t, compare it with the threshold according to the value of t, mark the partitions according to the comparison results, obtain the partition results of each map grid, map them on the construction site plane map, and obtain the final grid map state; Convert the final raster map state into vector boundaries, extract the boundaries of different partitions, represent different partitions with different colors on the construction site plan map, and mark the attributes of each partition on the construction site plan map.

2. The method for distributed scheduling and storage optimization of engineering resources based on access popularity according to claim 1, characterized in that: The specific working steps of S2 are as follows: Through the Internet of Things technology, prefabricated component manufacturers, transportation vehicles and warehouse nodes are interconnected to establish a real-time data channel; The manufacturer node is responsible for providing production progress data, inventory data, and quality inspection data for prefabricated components; The transport vehicle node is responsible for providing the vehicle's location, speed, type and quantity of loaded components, and estimated arrival time; The warehouse node is responsible for providing warehouse inventory data, storage location information, warehouse environment data, including temperature and humidity, and inbound and outbound record data.

3. The method for distributed scheduling and storage optimization of engineering resources based on access popularity according to claim 1, characterized in that: The S33 further comprises the following steps: When the stage label p is switched, the size of each partition on the construction site plan map is adjusted according to the maximum component access heat ratio between the current stage and the previous stage.

4. The method for distributed scheduling and storage optimization of engineering resources based on access popularity according to claim 1, characterized in that: The specific working steps of S4 are as follows: S41. Based on the access heat field and stress field Calculate the partition heat gradient difference and heat change rate; S42. Obtain a reference dimension, and automatically select a scheduling mode based on the comparison of the partition heat gradient difference multiplied by the heat change rate. If the partition heat gradient difference multiplied by the heat change rate is greater than the reference dimension, select the first scheduling mode; otherwise, select the second scheduling mode.

5. The method for distributed scheduling and storage optimization of engineering resources based on access popularity according to claim 4 is characterized in that: The step S4 further includes step S43, which is used to drive the scheduling execution according to the selected scheduling mode. The specific steps are as follows: If the first scheduling mode is used, the product of the visit popularity and the inventory ratio at each manufacturer is calculated, and the manufacturer with the largest product is selected. The visit popularity is the visit popularity at the manufacturer's location, and the inventory ratio is the manufacturer's inventory sufficiency, that is, the percentage of the current inventory divided by the maximum inventory; When the vehicle reaches the target partition and the time derivative of the access heat gradient is zero, unloading begins; If it is the second scheduling mode, first calculate the transportation frequency, during each transportation process; Find the point with the smallest access heat gradient in the heavy area as the unloading point; Calculate the partial derivative of the access heat in the height direction z at each component position in the vertical component area, and sort the components in descending order according to this value.

6. The method for distributed scheduling and storage optimization of engineering resources based on access popularity according to claim 1, characterized in that: The S5 comprises the following steps: S51. In the access heat field , calculate the ratio of planned access heat to actual access heat at each location, take the natural logarithm and perform weighted summation to obtain an overall difference metric; Obtain system parameters. When the overall difference metric is greater than the system parameters, it is determined that the project progress is lagging behind, and an alarm is triggered. Obtain the activity of the manufacturer's node. If the activity of the manufacturer's node is lower than the first quartile of the historical activity data, it is determined that the manufacturer's node has been interrupted, and an alarm is triggered; The stress field mutation rate is calculated by calculating the rate of change of the stress field gradient over time. If the stress field mutation rate exceeds the historical average plus two times the standard deviation, an alarm is triggered.

7. The method for distributed scheduling and storage optimization of engineering resources based on access popularity according to claim 6, characterized in that: The S5 also includes S52, which is used to re-evaluate the access popularity after receiving the alarm. The specific steps are as follows: After receiving the alarm, if the project progress is lagging behind, the updated heat value is calculated based on the overall difference measurement ; If it is the other two reasons of S51, the updated heat value is calculated based on the stress field mutation rate .

8. The method for distributed scheduling and storage optimization of engineering resources based on access popularity according to claim 7, characterized in that: The specific steps of S52 also include the following: The updated heat value Enter S3 to recalculate the access popularity field and trigger the reselection of the scheduling mode; Call S4 to reselect the scheduling mode and drive the scheduling execution according to the selected scheduling mode.

Images