Total group site layout and dynamic scheduling method for real-time feedback adjustment

Abstract

The invention discloses a total group site layout and dynamic scheduling method for real-time feedback adjustment. The method comprises the following steps: step 1, determining ship total group site resource elements and constraint conditions; 2, building a total group site real-time data acquisition and transmission system; 3, constructing a dynamic layout and scheduling model for real-time feedback; 4, generating an initial total group site layout and scheduling scheme; 5, real-time monitoring and data feedback are carried out; step 6, scheme evaluation and adjustment triggering; 7, generating and executing a dynamic adjustment scheme; 8, scheme filing and model optimization are carried out; according to the invention, the dynamic optimization of the layout and scheduling of the total group site is realized by collecting the dynamic data of the components, the equipment, the personnel and the stations in real time and combining the quantitative model and the multi-stage adjustment mechanism. According to the invention, the rigid constraint of a traditional fixed planning mode is broken through, millisecond-level data transmission is realized through the 5G and LoRa hybrid network, and real-time perception of a resource state is ensured.

Description

Technical Field

[0001] This invention relates to the field of shipbuilding engineering and resource optimization and management technology, specifically to a method for real-time feedback adjustment of the overall site layout and dynamic scheduling. Background Technology

[0002] Ship assembly is a core component of shipbuilding. The assembly site, as a resource-intensive work area, involves the coordination of multiple elements, including components, specialized equipment, skilled personnel, and workstations. Traditional ship assembly site layouts often employ a "fixed planning and experience-based scheduling" model, which presents the following problems:

[0003] The layout is rigid: the site workstations and equipment parking positions are fixed, which cannot respond to dynamic scenarios such as component delivery delays, temporary equipment failures, and process changes. This can easily lead to some workstations being overloaded while others are idle.

[0004] Scheduling lag: Relies on manual recording of resource status, resulting in long data feedback cycles and scheduling scheme adjustments failing to match real-time job requirements;

[0005] Insufficient coordination: The scheduling of components, equipment, and personnel is independent of each other, lacking a unified dynamic correlation model, which easily leads to resource waste such as "equipment waiting for components" and "personnel waiting for equipment";

[0006] Lack of quantitative evaluation mechanism: Without the establishment of real-time evaluation indicators for resource utilization and scheduling effectiveness, it is impossible to scientifically judge the rationality of scheduling schemes, adjustment decisions rely on experience, and the stability is poor.

[0007] Therefore, there is an urgent need for a real-time feedback adjustment method for the layout and dynamic scheduling of ship assembly sites. This method can solve the rigidity and lag problems of traditional methods by real-time data collection, quantitative model construction, and dynamic adjustment triggering and execution, thereby improving the resource utilization efficiency and operational coordination of ship assembly sites. Summary of the Invention

[0008] The purpose of this invention is to provide a method for real-time feedback adjustment of the overall site layout and dynamic scheduling, so as to solve the problems mentioned in the background art.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a method for real-time feedback adjustment of the overall site layout and dynamic scheduling, comprising the following steps:

[0010] Step 1: Determine the site resource elements and constraints of the ship assembly, sort out the parameter information of the four core resources of components, equipment, personnel and workstations, and clarify the four constraints of space, time, resources and safety;

[0011] Step 2: Build a real-time data acquisition and transmission system for the main site, deploy multiple types of sensors to collect resource status data, and construct a 5G+LoRa hybrid transmission network and data preprocessing module;

[0012] Step 3: Construct a dynamic layout and scheduling model with real-time feedback, including a spatial layout optimization sub-model (minimizing the total transport distance) and a resource scheduling optimization sub-model (introducing a resource load balancing index).

[0013] Step 4: Generate the initial overall site layout and scheduling plan, use an improved genetic algorithm to solve the model, and output a component-equipment-workstation matching table and a resource operation schedule. After review, the plan will be executed.

[0014] Step 5: Real-time monitoring and data feedback. Resource status data is continuously collected through the data acquisition system, analyzed on the cloud platform, and then fed back to the scheduling terminal.

[0015] Step Six: Scheme Evaluation and Adjustment Triggering. Every 15 minutes, the scheme is evaluated based on indicators such as resource load balancing and task progress deviation rate. Adjustment is triggered when the adjustment conditions are met.

[0016] Step 7: Dynamically adjust the generation and execution of the plan. Based on the adjustment level input parameters, introduce the task adjustment priority coefficient to determine the task order, generate the adjustment plan and track its execution.

[0017] Step 8: Project archiving and model optimization. Organize project and data archives, analyze data monthly and optimize model parameters.

[0018] Preferably, the resource elements in step one specifically include: the type, geometric parameters, weight, process requirements and delivery plan of the components; the type, technical parameters, operating status and scheduling authority of the equipment; the skill level, certificate type, on-the-job status and work efficiency benchmark of the personnel; the functional zoning, spatial parameters, supporting facilities and process adaptability of the workstation.

[0019] Preferably, the data preprocessing module in step two includes denoising, normalization, and correlation matching operations, wherein the normalization uses the formula:

[0020]

[0021] in, For the normalized data, This is the original data. The minimum value of this type of data. This represents the maximum value of this type of data.

[0022] Preferably, the resource load balancing degree in step three is expressed by the formula... calculate, For resource load balancing, For the first The actual resource load of each workstation. The average resource load for all workstations. This represents the total number of workstations.

[0023] Preferably, the improved genetic algorithm in step four includes encoding, population initialization, fitness function, and iterative optimization operation. The fitness function aims at "minimum total transport distance and maximum resource load balance" and iterates for 50-100 generations until convergence.

[0024] Preferably, the evaluation metrics in step six include resource load balancing. Task progress deviation rate Equipment utilization rate Personnel utilization rate The triggering conditions have been adjusted to include a single indicator exceeding the threshold twice consecutively, any two indicators exceeding the threshold simultaneously, or a major anomaly occurring.

[0025] Preferably, the task adjustment priority coefficient in step seven is calculated using the formula... calculate, For the first The priority coefficient of each task to be adjusted. The task delay risk factor. For resource dependence, Due to the urgency of the process, , , The weighting coefficients are satisfied. .

[0026] Preferably, the sensors in step two include component RFID tags, equipment GPS positioning modules, personnel smart work cards, and workstation infrared sensors, with the data acquisition frequency set according to the resource type.

[0027] Preferably, the adjustment levels in step seven are divided into three levels:

[0028] Level 1 adjustment, a partial adjustment, affecting one workstation or one piece of equipment: decided by the on-site dispatcher;

[0029] Level 2 adjustments, regional adjustments, affecting 2-3 workstations or 2 or more pieces of equipment: decided by the dispatch supervisor;

[0030] Level 3 adjustments, or global adjustments, affecting 4 or more workstations or core equipment: are decided by the chief dispatcher.

[0031] Preferably, the model optimization in step eight includes adjusting the number of iterations of the genetic algorithm, adjusting the weight coefficients of the task priority coefficients, and optimizing based on the monthly archived schemes and data statistical analysis results, forming an execution-feedback-optimization closed loop.

[0032] Compared with the prior art, the beneficial effects of the present invention are:

[0033] This invention achieves dynamic optimization of overall site layout and scheduling by collecting real-time dynamic data on components, equipment, personnel, and workstations, combined with a quantitative model and a multi-level adjustment mechanism. It breaks through the rigid constraints of traditional fixed planning models, achieving millisecond-level data transmission through a hybrid 5G and LoRa network to ensure real-time resource status awareness; it introduces resource load balancing and task priority coefficients to solve the problem of resource waste in multi-element collaborative scheduling; and it establishes a 15-minute cycle evaluation-trigger mechanism to improve the response speed of scheduling scheme adjustments. Attached Figure Description

[0034] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0035] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0036] Please see Figure 1 This invention provides a method for real-time feedback adjustment of the overall site layout and dynamic scheduling, comprising the following steps:

[0037] Step 1: Determine the site resources and constraints for ship assembly.

[0038] 1.1 Identify the core resource elements of the assembly site, specifically including: Component resources: Define the types of components required for assembly (e.g., compartments, deck blocks, superstructure), geometric parameters (length, width, height), weight, process requirements (e.g., welding sequence, pre-painting condition), and delivery schedule (planned delivery time, quantity); Equipment resources: Compile statistics on the types of specialized equipment on site (gantry cranes, tire cranes, flatbed trucks, welding robots), technical parameters (rated load, operating radius, moving speed), operating status (normal, maintenance, fault), and scheduling authority. Can cross-regional deployment be made? Personnel resources: Record the skill level of the operators (e.g., welders are divided into primary / intermediate / advanced), certificate type (e.g., special equipment operation certificate, welding qualification certificate), on-duty status (attendance, leave, time off) and work efficiency benchmark (number of processes completed per unit time); Workstation resources: Divide the functional areas of the workstations in the site (general assembly area, pre-assembly area, storage area, maintenance area), spatial parameters (area, load-bearing capacity, width of access passage), supporting facilities (e.g., power interface, lifting equipment anchor point) and process adaptability (whether it can meet the needs of large component assembly).

[0039] 1.2 Define the constraints in the assembly process, including: Spatial constraints: The workstation area must match the geometric dimensions of the components, and the equipment operating radius must cover the component hoisting range to avoid conflicts in equipment movement trajectories between different workstations; Time constraints: The component assembly process must follow the process sequence (e.g., complete the module docking before welding), and equipment scheduling must meet the process time nodes (e.g., the crane must be in place within 1 hour of the component arrival); Resource constraints: The same equipment cannot be assigned to multiple tasks simultaneously, and highly skilled personnel should be prioritized for high-difficulty processes (e.g., thick plate welding); Safety constraints: The equipment operating speed must meet site safety standards, and the component storage spacing must meet fire prevention and explosion-proof requirements (e.g., the storage spacing of cryogenic components for LNG ships is ≥5 meters).

[0040] Step 2: Establish a real-time data acquisition and transmission system for the main venue.

[0041] 2.1 Deploy multiple types of sensors and data acquisition terminals, specifically including: Component status acquisition: RFID tags are affixed to the surface of components, and RFID readers deployed on-site collect the component location (accuracy down to the meter level) and storage time in real time; temperature and humidity sensors are installed in the component storage area (for special components such as cryogenic storage tanks) to collect environmental parameters; Equipment status acquisition: GPS positioning modules are installed on equipment (cranes, transport vehicles) to collect equipment location in real time; current, voltage, and speed sensors are connected to the equipment controller to collect equipment operating parameters (such as crane lifting capacity and transport vehicle speed); status indicator lights are installed on the equipment operation panel to collect equipment working status (running, standby, fault); Personnel status acquisition: Smart work cards are provided to operators, and personnel location is collected in real time via Bluetooth base stations on-site; the attendance module built into the work card records personnel on-duty / off-duty status; Workstation status acquisition: Infrared sensors are installed around the workstation to detect whether the components in the workstation are in place; power sensors are installed in the workstation power distribution box to collect workstation energy consumption (reflecting work intensity).

[0042] 2.2 Construction of Data Transmission and Preprocessing Modules: Data Transmission: A hybrid transmission network of 5G industrial Ethernet and LoRa is adopted, with 5G used for high-frequency data transmission and LoRa used for low-frequency data transmission to ensure data real-time performance and stability; Data Preprocessing: The raw data collected is processed through edge computing nodes, including: Noise Reduction: Abnormal data (such as device position jumps caused by instantaneous sensor fluctuations) is removed using the moving average method; Normalization: Data of different dimensions is converted to the [0,1] interval, using the following formula:

[0043]

[0044] in, For the normalized data, This is the original data. The minimum value of this type of data. This represents the maximum value for this type of data; Association matching: Associate component ID, equipment ID, personnel ID, and workstation ID to form a "resource-workstation" correspondence table to ensure data traceability.

[0045] Step 3: Construct a dynamic layout and scheduling model with real-time feedback.

[0046] 3.1 Establishing a spatial layout optimization sub-model: Objective: To achieve optimal matching between workstations and components / equipment, minimizing resource handling distances between workstations; Variable definition: Let... For 0-1 variables, Indicates the first Class of components ( ) through the first Class of equipment ( ) allocated to the workstations ( ), =0, then the opposite; let... For the first The workstation and the first The transport distance between workstations; constraints: each component is assigned to only one workstation, the total weight of components at each workstation does not exceed its load-bearing capacity, and the number of devices at each workstation does not exceed the number of its associated device interfaces; objective function: minimize the total transport distance. .

[0047] 3.2 Establish a resource scheduling optimization sub-model: Objective: To achieve real-time matching of equipment, personnel, and tasks, maximizing resource utilization; introduce resource load balancing. As a core evaluation indicator, it is used to quantify the balance of resource load across workstations. The formula is:

[0048]

[0049] in: The resource load balancing degree has a value range of [0,1]. The closer it is to 1, the more balanced the load is across all workstations; For the first The actual resource load of each workstation (taking into account equipment operating time, personnel working time, and component area, calculated by weighting the normalized data in step 2.2, with weights set according to process importance). The average resource load for all workstations. ; This represents the total number of workstations.

[0050] Constraints: Equipment scheduling routes must not violate site traffic rules (such as one-way passages), personnel working hours must not exceed the daily rated working hours, and task completion time must not exceed the deadline required by the process.

[0051] Step 4: Generate the initial overall site layout and scheduling plan

[0052] 4.1 Input basic data: Input the resource elements (component list, equipment parameters, personnel information, workstation parameters) and constraints (space, time, safety constraints) determined in step 1 into the model constructed in step 3;

[0053] 4.2 Model Solving: An improved genetic algorithm is used to solve the model, specifically including: Encoding: Binary encoding is used, and the chromosome length is... (correspond (Quantity), each representing a The value of ; Initialize the population: randomly generate 100-200 initial chromosomes, ensuring that each chromosome meets the constraints; Fitness function: use "minimum total transport distance D and maximum resource load balance L" as the fitness function. The higher the fitness value, the better the solution; Iterative optimization: through selection (roulette wheel), crossover (single-point crossover), and mutation (random mutation) operations, iterate for 50-100 generations until the fitness value converges, and output the initial layout scheme (component-equipment-workstation matching table) and scheduling scheme (equipment operation timetable, personnel shift schedule).

[0054] 4.3 Scheme Review: Organize process, safety, and scheduling personnel to review the initial scheme, focusing on whether it meets safety constraints (such as no conflict in equipment movement trajectory) and process constraints (such as correct process sequence). If the review is passed, proceed to the execution stage; if it fails, return to step 4.2 to adjust the model parameters and solve again.

[0055] Step 5: Real-time monitoring and data feedback

[0056] 5.1 Real-time data acquisition: Through the system built in step 2, continuously collect data on component status (location, storage duration), equipment status (location, operating parameters, fault information), personnel status (location, on-duty time), and workstation status (load, energy consumption). The acquisition frequency is set according to the resource type.

[0057] 5.2 Real-time Data Analysis: The collected data is analyzed in real time on the cloud-based management platform, specifically including: Resource Status Analysis: Calculating the real-time load of each workstation. Compared with the initial plan The system includes: load deviation assessment; task progress analysis: based on the component assembly progress (calculated by welder attendance records and equipment operation time), comparing it with the planned progress to determine the progress deviation; anomaly identification: setting an anomaly threshold, triggering an anomaly alarm when the data exceeds the threshold, and recording the anomaly type, occurrence time, and associated resource ID.

[0058] 5.3 Data Feedback: Real-time analysis results (load deviation, schedule deviation, and abnormal information) are fed back to the dispatch terminal through the cloud platform, with a feedback delay of no more than 30 seconds, to ensure that dispatchers can keep track of the site status in real time.

[0059] Step Six: Triggering the Evaluation and Adjustment of the Solution

[0060] 6.1 Dynamic Evaluation: A quantitative evaluation of the current layout and scheduling scheme is conducted every 15 minutes. Evaluation indicators include: resource load balancing. Based on the real-time load calculated in step 5.2 The formula in step 3.2 is used for calculation. Task progress deviation rate : ,in This represents the actual progress (e.g., 30% complete). For planned progress (e.g., 40% complete); equipment utilization rate : ,in This refers to the actual operating time of the equipment. Planned operating time of equipment; personnel utilization rate : ,in For actual working hours of personnel, The duration of staff on duty.

[0061] 6.2 Adjust Trigger Judgment: Set evaluation indicator thresholds (e.g.) <0.7 >10%, <60%, <70%), the scheme adjustment is triggered when any of the following conditions are met: a single indicator exceeds the threshold twice consecutively (e.g., Two consecutive values ​​of 0.65 < 0.7; any two indicators exceeding the threshold simultaneously (e.g., =12%>10% and =55%<60%); major anomalies occur, such as main crane failure or delays in the delivery of components exceeding 4 hours.

[0062] 6.3 Adjustment Level Classification: Based on the triggering cause and the scope of impact, adjustments are divided into three levels: Level 1 Adjustment (Local Adjustment): Affects 1 workstation or 1 piece of equipment (e.g., workstation 3 has excessive load), decided by the on-site dispatcher; Level 2 Adjustment (Regional Adjustment): Affects 2-3 workstations or 2 or more pieces of equipment (e.g., equipment failure in a certain area), decided by the dispatch supervisor; Level 3 Adjustment (Global Adjustment): Affects 4 or more workstations or core equipment (e.g., main crane failure), decided by the overall dispatch manager.

[0063] Step 7: Dynamically Adjusting the Generation and Execution of the Plan

[0064] 7.1 Adjustment Parameter Input: Based on the adjustment level, input the relevant adjustment parameters, including: abnormal resource information (such as faulty equipment ID, overloaded workstation ID); available backup resources (such as backup crane ID, idle workstation ID); adjustment constraints (such as the task progress deviation after adjustment must be ≤5%, load balancing degree). ≥0.75).

[0065] 7.2 Introduce a task priority adjustment coefficient This is used to determine the priority order of tasks to be adjusted, and the formula is:

[0066]

[0067] in: For the first The priority coefficient of each task to be adjusted, with a value range of [0,1]. The larger the value, the higher the task priority; This represents the risk factor for task delay; the greater the deviation in the current task progress, the higher the risk factor. The closer it is to 1, such as a schedule deviation of 12%. =0.9; Resource dependency: The scarcer the resources a task depends on, the higher the resource dependency. The closer it is to 1, the more important it is to have a task that relies on the main crane. =0.95; Based on the urgency of the process, the earlier the task appears in the overall workflow, the better. The closer it is to 1, the better, such as in a module docking mission. =0.9; , , For the weighting coefficients, satisfying The settings are based on process requirements, such as prioritizing critical processes. =0.4, =0.3, =0.3.

[0068] 7.3 Adjustment Plan Generation: Level 1 Adjustment: Using a local search algorithm, only the component allocation of overloaded workstations is adjusted (e.g., transferring one component from workstation 3 to workstation 5), and the calculation is recalculated. until ≥0.75, generate adjustment schemes (component transfer instructions, equipment scheduling instructions); Secondary adjustment: use a genetic algorithm for local iteration (20-30 generations) to adjust the equipment-task allocation within the region (e.g., assigning tasks from faulty equipment to standby equipment) to ensure the region's... ≥0.75 and ≤5%, generate adjustment plan; Level 3 adjustment: return to step 4.2, re-enter the adjusted resource elements (such as removing faulty equipment), and use the improved genetic algorithm to solve again (iterate 50-80 generations) to generate a global adjustment plan.

[0069] 7.4 Implementation and Tracking of the Plan: Instruction Issuance: The adjustment plan (equipment scheduling instructions, personnel adjustment instructions, component transfer instructions) is issued to relevant personnel (such as equipment operators and crane operators) via the dispatch terminal. Instructions must include execution time, operating steps, and safety requirements; Execution Tracking: The execution progress of the adjustment plan is tracked through a real-time data acquisition system (e.g., whether equipment moves according to instructions, whether components are transferred to the target workstation), with the execution status updated every 5 minutes; Execution Verification: After the adjustment plan is completed, the evaluation indicators are recalculated. , , , If the indicator meets the threshold requirement, the adjustment ends; otherwise, return to step 7.2 to regenerate the adjustment plan.

[0070] Step 8: Solution Archiving and Model Optimization

[0071] 8.1 Solution Archiving: Organize the initial solution, adjusted solutions (including triggering reasons, adjusted parameters, and execution results), and real-time data (including anomaly records) into archives and store them in the cloud database. The archives must include the solution ID, execution time period, and associated general group project ID to facilitate subsequent traceability and analysis.

[0072] 8.2 Model Optimization: Monthly statistical analysis of archived schemes and data, specifically including: adjusting trigger cause statistics (e.g., equipment failure rate, overload rate); model parameters (e.g., genetic algorithm iteration count, weight coefficients). , , Impact analysis on the effectiveness of the solution; adjust model parameters based on the analysis results (e.g., increase parameters if equipment failures are frequent). weight This optimizes the model's solution efficiency and the adaptability of the solution, forming a closed loop of "execution-feedback-optimization".

[0073] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for real-time feedback adjustment of total group venue layout and dynamic scheduling, characterized in that: Includes the following steps: Step 1: Determine the site resource elements and constraints of the ship assembly, sort out the parameter information of the four core resources of components, equipment, personnel and workstations, and clarify the four constraints of space, time, resources and safety; Step 2: Build a real-time data acquisition and transmission system for the main site, deploy multiple types of sensors to collect resource status data, and construct a 5G+LoRa hybrid transmission network and data preprocessing module; Step 3: Construct a dynamic layout and scheduling model with real-time feedback, including a spatial layout optimization sub-model and a resource scheduling optimization sub-model; Step 4: Generate the initial overall site layout and scheduling plan, use an improved genetic algorithm to solve the model, and output a component-equipment-workstation matching table and a resource operation schedule. After review, the plan will be executed. Step 5: Real-time monitoring and data feedback. Resource status data is continuously collected through the data acquisition system, analyzed on the cloud platform, and then fed back to the scheduling terminal. Step Six: Scheme Evaluation and Adjustment Triggering. Every 15 minutes, the scheme is evaluated based on indicators such as resource load balancing and task progress deviation rate. Adjustment is triggered when the adjustment conditions are met. Step 7: Dynamically adjust the generation and execution of the plan. Based on the adjustment level input parameters, introduce the task adjustment priority coefficient to determine the task order, generate the adjustment plan and track its execution. Step 8: Project archiving and model optimization. Organize project and data archives, analyze data monthly and optimize model parameters.

2. The method of claim 1, wherein: The resource elements in step one specifically include: the type, geometric parameters, weight, process requirements and delivery plan of the components; the type, technical parameters, operating status and scheduling authority of the equipment; the skill level, certification type, on-the-job status and work efficiency benchmark of the personnel; and the functional zoning, spatial parameters, supporting facilities and process adaptability of the workstations.

3. The method of claim 1, wherein: The data preprocessing module in step two includes noise reduction, normalization, and correlation matching operations, wherein the normalization uses the following formula: ； wherein, is the normalized data, is the original data, is the minimum value of the data of this class, is the maximum value of the data of this class.

4. The method for real-time feedback adjustment of the overall site layout and dynamic scheduling according to claim 1, characterized in that: The resource load balancing degree in the step three adopts the formula Calculate, The resource load balancing degree, The actual resource load of the first The actual resource load of the first The average resource load of all workstations, The total number of workstations.

5. The method of claim 1, wherein: The improved genetic algorithm in step four includes encoding, population initialization, fitness function, and iterative optimization. The fitness function aims at "minimum total transport distance and maximum resource load balance", and iterates for 50-100 generations until convergence.

6. The method for real-time feedback adjustment of the overall site layout and dynamic scheduling according to claim 1, characterized in that: The evaluation index in the step six includes resource load balancing degree , task progress deviation rate , equipment utilization rate , personnel utilization rate The adjustment trigger condition is that a single index exceeds the threshold value for two times continuously, any two indexes exceed the threshold value at the same time, or a major abnormality occurs.

7. The method for real-time feedback adjustment of the overall site layout and dynamic scheduling according to claim 1, characterized in that: The task adjustment priority coefficient in step seven is calculated using the formula... calculate, For the first The priority coefficient of each task to be adjusted. The task delay risk factor. For resource dependence, Due to the urgency of the process, , , The weighting coefficients are and satisfy the following conditions: .

8. The method for real-time feedback adjustment of the overall site layout and dynamic scheduling according to claim 1, characterized in that: The sensors in step two include component RFID tags, equipment GPS positioning modules, personnel smart work cards, and workstation infrared sensors. The data acquisition frequency is set according to the resource type.

9. The method for real-time feedback adjustment of the overall site layout and dynamic scheduling according to claim 1, characterized in that: The adjustment levels in step seven are divided into three levels: Level 1 adjustment, a partial adjustment, affecting one workstation or one piece of equipment: decided by the on-site dispatcher; Level 2 adjustments, regional adjustments, affecting 2-3 workstations or 2 or more pieces of equipment: decided by the dispatch supervisor; Level 3 adjustments, or global adjustments, affecting 4 or more workstations or core equipment: are decided by the chief dispatcher.

10. The method for real-time feedback adjustment of the overall site layout and dynamic scheduling according to claim 1, characterized in that: The model optimization in step eight includes adjusting the number of iterations of the genetic algorithm, adjusting the weight coefficients of the task priority coefficients, and optimizing based on the monthly archived schemes and data statistical analysis results, forming an execution-feedback-optimization closed loop.

Images