Equipment operation and maintenance decision-making method and system, terminal and medium

Abstract

The invention belongs to the technical field of equipment operation and maintenance, and particularly discloses an equipment operation and maintenance decision-making method and system, a terminal and a medium. Comprising the following steps: acquiring equipment operation data, operation and maintenance personnel state data, spare part resource data and position service data, and preprocessing the data to obtain a structured data set; equipment state prediction processing is executed based on the equipment operation data, and a prediction result used for representing the future operation state of the equipment is obtained; calculating cost information based on the position service data, the state data and the spare part resource data to form operation and maintenance cost data; and inputting the prediction result and the operation and maintenance cost data into an operation and maintenance decision model, solving under personnel allocation constraint, task sequence constraint and path planning constraint, and generating operation and maintenance decision data including personnel allocation, task sequence, travel path and spare part distribution. By providing a unified cost quantification model and a dynamic adjustment capability based on real-time data, cost optimization and adaptive adjustment of an execution process can be considered at the same time.

Description

Technical Field

[0001] This invention belongs to the field of equipment operation and maintenance technology, and specifically relates to an equipment operation and maintenance decision-making method, system, terminal and medium. Background Technology

[0002] In industries such as power, energy, manufacturing, and transportation, the daily inspection, maintenance, and emergency repair of large-scale equipment are characterized by high frequency, wide geographical coverage, and large workload. Traditional methods relying on manual scheduling are no longer sufficient to meet the real-time and refined requirements of modern operation and maintenance. As the level of digitalization in operation and maintenance continues to increase, equipment operation data, personnel status data, spare parts resource data, and location service data are gradually becoming important foundations for operation and maintenance management.

[0003] In existing technologies, common operation and maintenance decision-making processes typically include: collecting equipment operation data based on monitoring platforms, using empirical rules or fixed thresholds to determine whether equipment needs maintenance; generating maintenance tasks using a work order system, and assigning task order to personnel based on their skill information, availability information, and geographical location, combined with route planning algorithms; in terms of path planning, existing technologies mostly adopt static route calculation methods, determining the travel distance and estimated arrival time based on geographical location to support task sorting and execution.

[0004] However, existing technologies still have certain problems: existing scheduling systems lack a unified cost quantification model and dynamic adjustment capabilities based on real-time data, and cannot simultaneously take into account cost optimization and adaptive adjustment of the execution process. Summary of the Invention

[0005] This invention addresses the problems in the prior art by providing a method, system, terminal, and medium for equipment operation and maintenance decision-making. This solves the problem in the prior art where scheduling systems lack a unified cost quantification model and dynamic adjustment capabilities based on real-time data, making it impossible to simultaneously achieve cost optimization and adaptive adjustment of the execution process.

[0006] The technical solution adopted in this invention is as follows: Firstly, this application provides a method for making equipment operation and maintenance decisions, which includes the following steps: Acquire initial data, which includes equipment operation data, maintenance personnel status data, spare parts resource data, and location service data. Preprocess the initial data to obtain a structured data set. Based on the equipment's operating data, perform equipment status prediction processing to obtain prediction results that reflect the future operating status of the equipment; Based on location service data, status data and spare parts resource data, the cost elements in the operation and maintenance process are quantified, and the cost elements are summarized one by one according to the preset cost accounting model to generate the corresponding operation and maintenance cost data. The prediction results and operation and maintenance cost data are input into the operation and maintenance decision model. Based on the objective function, constraints and optimization rules preset by the operation and maintenance decision model, the optimal or suboptimal solution that meets the conditions is searched in the solution space to obtain operation and maintenance decision data that includes personnel allocation information, task sequence information, travel path information and spare parts delivery information.

[0007] Furthermore, in the pre-trained equipment state prediction model, equipment state prediction processing is performed based on the equipment's operating data. The training of the equipment state prediction model includes the following steps: Acquire sample data containing historical operating parameters and historical fault records. Historical operating parameters are used to characterize the operating status of the equipment at different times, and historical fault records are used to indicate whether the equipment has experienced a fault or the time and location of the fault within the corresponding time period. Based on a preset time window, the historical operating parameters are divided into multiple consecutive operating state sequences in chronological order. Each operating state sequence is combined with its corresponding historical fault record to form a sample pair for training. A device status prediction model is established based on sample pairs, and the model is trained by comparing the correspondence between historical operating status and historical fault records. After the model training is completed, the sequence of the target device's operating state before the current moment is input into the device state prediction model to obtain the prediction results used to characterize the future operating state of the target device. The prediction results are the probability information of the device's failure and / or remaining operating life information within the preset prediction time range.

[0008] Furthermore, the calculation of operation and maintenance costs based on location service data, status data, and spare parts resource data includes: Based on location service data, determine the travel distance information and corresponding travel time information between each maintenance personnel and each target device; Based on equipment status data and maintenance task plan data, determine the on-site operation time and equipment downtime information for each maintenance task; Based on spare parts resource data, determine the inventory status of spare parts required for each maintenance task, the spare parts arrival time, and the waiting time caused by the spare parts arrival time, and correct the corresponding equipment downtime information. The number of business trip days corresponding to each operation and maintenance task is determined based on calendar date information and start and end time information of operation and maintenance tasks. The unit price parameters for transportation costs, accommodation costs, catering costs, subsidy costs, labor costs, and equipment downtime losses are read from the cost parameter configuration. The travel distance information, business trip days information, on-site operation time information, and equipment downtime time information are correlated with the corresponding unit price parameters to form operation and maintenance cost calculation data that characterizes the transportation costs, accommodation costs, catering costs, subsidy costs, labor costs, and equipment downtime losses of each operation and maintenance task.

[0009] Furthermore, the total maintenance cost corresponding to the maintenance cost calculation data is calculated according to the following formula:

[0010] in, Let be the travel distance of the i-th maintenance task; Let be the number of business trip days corresponding to the i-th maintenance task; Let be the on-site operation time for the i-th maintenance task; Let i be the equipment downtime corresponding to the i-th maintenance task; The waiting time for the i-th maintenance task caused by the spare parts delivery time; This represents the probability of the i-th device failing within a preset prediction time range, based on the device status prediction model. , , , , , These are the unit price parameters for transportation costs, accommodation costs, catering costs, subsidies, labor costs, and equipment downtime losses, respectively. This is a parameter sensitive to downtime risk; This is a parameter sensitive to waiting time.

[0011] Furthermore, the prediction and calculation results are input into the operation and maintenance decision model, and the operation and maintenance decision model is solved, including: Task urgency parameters are constructed based on the prediction results of the future operating status of the equipment, and cost constraint parameters are constructed based on the operation and maintenance cost calculation data. The task urgency parameters and cost constraint parameters are used as input elements of the operation and maintenance decision model. Based on the availability information of maintenance personnel, the order of task execution, and the accessibility information of travel paths, personnel allocation constraints, task sequencing constraints, and path planning constraints are constructed. Input elements and constraints are input into the operation and maintenance decision model, and iterative solutions are performed on the operation and maintenance decision model to obtain a set of feasible solutions that satisfy the constraints. Select the target decision solution from the set of feasible solutions, which includes personnel allocation information, task sequence information, travel path information, and spare parts delivery information, as the operation and maintenance decision data.

[0012] Furthermore, it also includes: During the execution of operation and maintenance tasks, real-time data is acquired to characterize the personnel execution status, spare parts delivery status, and equipment operation status. Executed or determined scheduling operations are set as fixed constraints, and the unexecuted scheduling parts are adjusted and calculated based on the updated operation and maintenance cost data expression to obtain updated operation and maintenance decision data.

[0013] Furthermore, the adjustment calculation for the unexecuted scheduling portion based on the updated operation and maintenance cost data expression includes: Based on real-time data on personnel execution status, the current location, current task progress, and available execution time window of the target personnel are updated to form real-time constraint information that characterizes personnel availability. Based on real-time data on spare parts delivery status, the location of spare parts en route, estimated arrival time, and possible delay time are updated to form real-time constraint information to characterize material availability. Based on real-time data on equipment operating status, the operating trend, risk level changes, and allowable downtime window of the target equipment are updated to form real-time constraint information that characterizes the equipment demand status. The real-time constraints on personnel availability, material availability, and equipment demand, along with the updated operation and maintenance cost data expression, are input into the operation and maintenance decision model. The unexecuted scheduling parts are then re-solved to obtain a feasible scheduling solution that meets the real-time constraints. Select the scheduling result that is compatible with the original operation and maintenance decision data from the feasible scheduling solutions, and replace the corresponding unexecuted part in the original operation and maintenance decision data with the scheduling result to form the updated operation and maintenance decision data.

[0014] Secondly, this application provides an equipment operation and maintenance decision-making system for implementing the equipment operation and maintenance decision-making method as described in the first aspect. The system includes: The data acquisition and preprocessing unit is used to acquire equipment operation data, maintenance personnel status data, spare parts resource data, and location service data, and to perform formatting and data alignment processing on the data to obtain a structured data set. The equipment status prediction unit is used to perform equipment status prediction processing based on equipment operation data in a structured dataset, and output prediction results to characterize the future operating status of the equipment. The operation and maintenance cost calculation unit is used to calculate the travel distance information, business trip days information, operation duration information and equipment downtime information of each operation and maintenance task based on location service data, status data and spare parts resource data, and generate operation and maintenance cost calculation data and total operation and maintenance cost according to cost parameter configuration. The operation and maintenance decision solving unit is used to input the prediction results output by the equipment status prediction unit and the operation and maintenance cost calculation data output by the operation and maintenance cost calculation unit into the operation and maintenance decision model, and solve it under personnel allocation constraints, task sequencing constraints and path planning constraints to obtain operation and maintenance decision data containing personnel allocation information, task sequence information, travel path information and spare parts delivery information. The dynamic adjustment unit for operation and maintenance is used to acquire real-time data on personnel execution status, spare parts delivery status, and equipment operation status during the execution of operation and maintenance tasks. It sets executed or determined scheduling operations as fixed constraints, and adjusts and calculates the unexecuted scheduling parts based on the updated operation and maintenance cost data to obtain updated operation and maintenance decision data.

[0015] Thirdly, this application provides a terminal, including: Memory, used to store device operation and maintenance decision-making programs; A processor is used to implement the steps of the equipment operation and maintenance decision-making method as described in the first aspect when executing the equipment operation and maintenance decision-making device.

[0016] Fourthly, this application provides a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the device operation and maintenance decision-making method as described in the first aspect.

[0017] As can be seen from the above technical solutions, the advantages of the present invention are: By uniformly acquiring and preprocessing equipment operation data, personnel status data, spare parts resource data, and location service data, the structured integration of multi-source heterogeneous operation and maintenance data is achieved, enabling subsequent prediction, calculation, and scheduling to be carried out on a unified data basis, thereby improving the integrity and reliability of the decision-making chain.

[0018] By constructing a device status prediction model and training it based on historical operating parameters and fault records, a forward-looking assessment of the future operating status of the target device can be achieved. This enables the scientific quantification of task urgency, provides more accurate risk input for the scheduling model, and improves the rationality of operation and maintenance task sequencing and resource allocation.

[0019] By extracting key elements such as travel distance, business trip days, operation duration and downtime from location service data, status data and spare parts resource data, and combining them with cost parameters to configure execution cost quantification, the operation and maintenance cost calculation is transformed from experience-driven to quantifiable modeling, enabling scheduling decisions to be optimized with cost optimization as the goal.

[0020] By constructing a total cost expression that includes risk-sensitive factors and waiting time-sensitive factors, a unified quantification of transportation costs, accommodation costs, labor costs, and downtime losses can be achieved. This enables cost calculation results to reflect dynamic factors such as changes in equipment risk and the timeliness of spare parts, thereby improving the practical adaptability and decision-making value of cost evaluation.

[0021] By inputting the prediction results, cost data, personnel availability, task priority, and path reachability into the operation and maintenance decision model, and performing iterative solutions under constraints, the comprehensive optimization of personnel allocation, task order, and path planning is achieved, so that the scheduling results can simultaneously meet the requirements of urgency, feasibility, and economy.

[0022] By introducing real-time data on personnel execution status, spare parts delivery status, and equipment operation status during task execution, and recalculating the unexecuted parts after the executed parts are locked, the scheduling scheme can be dynamically updated, enabling the scheme to maintain feasibility and optimality as the execution environment changes, thereby improving the adaptive capability of the scheduling system.

[0023] By updating the real-time constraint information of personnel, materials and equipment, and resolving the unexecuted tasks in the operation and maintenance decision model, a smooth connection between real-time scheduling updates and the original plan is achieved, enabling operation and maintenance decisions to maintain continuous, stable and optimized execution under complex field conditions. Attached Figure Description

[0024] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart illustrating the steps of the equipment operation and maintenance decision-making method in the embodiment; Figure 2 This is a data update diagram of the equipment operation and maintenance decision-making method in the embodiment; Figure 3 This is a flowchart of the equipment operation and maintenance decision-making system in the embodiment. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] Please see Figure 1and Figure 2 As shown, this application provides a method for equipment operation and maintenance decision-making, including: Step S1: Obtain initial data, which includes equipment operation data, maintenance personnel status data, spare parts resource data, and location service data. Preprocess the initial data to obtain a structured data set. In practical implementation, equipment operation data can be automatically synchronized by the equipment's own sensing components, monitoring platforms, or remote acquisition systems, including real-time data characterizing equipment health and workload. Maintenance personnel status data can be reported via mobile terminals, including current location, work status, available time periods, and skill attributes, supporting subsequent task assignment. Spare parts resource data can be obtained through warehousing or supply chain management systems, including spare parts inventory, model matching relationships, supplier information, and estimated delivery times. Location service data can be collected through map service platforms or GPS terminals to calculate the spatial relationship between personnel and equipment. To ensure format consistency and time alignment between data from different sources, timestamp correction, encoding normalization, missing value completion, and data noise processing can be performed on the collected data, giving the structured data set a unified data structure and attributes directly usable for prediction and calculation. In a typical scenario, the system continuously receives various types of data through interfaces and automatically organizes them into task-oriented structured data entries before entering the decision-making process.

[0028] Step S2: Perform equipment status prediction processing based on the equipment's operating data to obtain prediction results that reflect the future operating status of the equipment; In practical implementation, equipment operation data can be divided into continuous time segments to identify equipment operation trends, periodic changes, and abnormal patterns. Historical operation data and historical failure records can be used to train predictive models, enabling them to learn the correspondence between equipment status and anomalies from indicators such as equipment temperature, vibration, current, pressure, and operating frequency. In practical applications, the system can extract key attributes reflecting equipment degradation trends from different types of data through automated data cleaning and feature extraction. During the prediction phase, the system generates predictions of potential risks in future time periods based on the recent operating conditions of the target equipment, and provides time trends indicating possible equipment anomalies. For equipment that is continuously under high load or experiences frequent operational fluctuations, the prediction results can reflect potential risks in advance, allowing the system to rationally prioritize tasks during the scheduling phase and providing effective input for subsequent cost calculations and resource allocation.

[0029] Step S3: Based on location service data, status data and spare parts resource data, quantify the cost elements in the operation and maintenance process, summarize the cost elements one by one according to the preset cost accounting model, and generate the corresponding operation and maintenance cost data. In practical implementation, location service data can be used to calculate the travel path of maintenance personnel from their current location to the target equipment location, thereby estimating the travel distance and corresponding travel time. Combining the personnel's current status and the equipment task plan, the expected duration of on-site operations can be calculated, and the impact of downtime can be estimated based on the equipment downtime window. Spare parts resource data is used to determine whether tasks will incur additional time losses due to waiting for spare parts, and related costs are adjusted based on spare parts delivery progress. Regarding cost parameter configuration, the system can read unit prices for expenses such as transportation, accommodation, catering, subsidies, labor, and downtime losses from the maintenance management platform, enabling the cost structure of different types of tasks to be quantified in a unified manner. The final calculated maintenance cost data can be used for comparison between different tasks, allowing scheduling optimization to aim for cost minimization. In an exemplary application, after multiple tasks are entered, the system automatically generates a cost estimate for each task to support the selection of feasible solutions for subsequent scheduling algorithms.

[0030] Step S4: Input the prediction results and operation and maintenance cost data into the operation and maintenance decision model. Based on the objective function, constraints and optimization rules preset by the operation and maintenance decision model, search for the optimal or suboptimal solution that meets the conditions in the solution space to obtain operation and maintenance decision data that includes personnel allocation information, task sequence information, travel path information and spare parts delivery information.

[0031] In practical systems, the operation and maintenance decision-making model can comprehensively process multiple types of inputs, including constraints such as task urgency, personnel availability, geographical accessibility, and spare parts delivery time, in order to identify executable scheduling combinations among different alternatives. During the solution process, the model can progressively check personnel skill matching, dependencies between tasks, and accessibility between different paths, generating feasible scheduling schemes while meeting constraints. For complex scenarios involving multiple tasks and personnel, the model can internally maintain multiple candidate solutions and progressively filter them to ensure that the final output decision meets actual requirements in terms of execution order, path connectivity, and spare parts usage. In typical implementations, when multiple tasks are distributed across different regions, the system can prioritize high-risk tasks based on equipment risk prediction results and appropriately insert other tasks between personnel schedules, thereby reducing waiting time and improving overall resource utilization.

[0032] In some embodiments, device state prediction processing is performed based on the device's operating data in a pre-trained device state prediction model. The training of the device state prediction model includes the following steps: Acquire sample data containing historical operating parameters and historical fault records. Historical operating parameters are used to characterize the operating status of the equipment at different times, and historical fault records are used to indicate whether the equipment has experienced a fault or the time and location of the fault within the corresponding time period. Based on a preset time window, the historical operating parameters are divided into multiple consecutive operating state sequences in chronological order. Each operating state sequence is combined with its corresponding historical fault record to form a sample pair for training. A device status prediction model is established based on sample pairs, and the model is trained by comparing the correspondence between historical operating status and historical fault records. After the model training is completed, the sequence of the target device's operating state before the current moment is input into the device state prediction model to obtain the prediction results used to characterize the future operating state of the target device. The prediction results are the probability information of the device's failure and / or remaining operating life information within the preset prediction time range.

[0033] In practical implementation, the system can select different historical data sources based on equipment type, operating environment, and monitoring frequency, including data automatically reported by online monitoring devices, management platforms, or handheld terminals. To improve training effectiveness, noise filtering and outlier correction can be performed on the data before training to make the training samples more stable and reliable. During the prediction phase, the model can generate future state judgments based on the recent operational fluctuation trends of the equipment, changes in key indicators, and historical degradation patterns, enabling the system to pre-mark high-risk equipment. In practical applications, for continuously operating power equipment, the model prediction results can be used to plan maintenance windows in advance, avoiding unexpected downtime during high-load periods.

[0034] In some embodiments, calculating operation and maintenance cost data based on location service data, status data, and spare parts resource data includes: Based on location service data, determine the travel distance information and corresponding travel time information between each maintenance personnel and each target device; Based on equipment status data and maintenance task plan data, determine the on-site operation time and equipment downtime information for each maintenance task; Based on spare parts resource data, determine the inventory status of spare parts required for each maintenance task, the spare parts arrival time, and the waiting time caused by the spare parts arrival time, and correct the corresponding equipment downtime information. The number of business trip days corresponding to each operation and maintenance task is determined based on calendar date information and start and end time information of operation and maintenance tasks. The unit price parameters for transportation costs, accommodation costs, catering costs, subsidy costs, labor costs, and equipment downtime losses are read from the cost parameter configuration. The travel distance information, business trip days information, on-site operation time information, and equipment downtime time information are correlated with the corresponding unit price parameters to form operation and maintenance cost calculation data that characterizes the transportation costs, accommodation costs, catering costs, subsidy costs, labor costs, and equipment downtime losses of each operation and maintenance task.

[0035] In practical implementation, the system can automatically retrieve route planning results from map services and combine them with real-time traffic data to generate more accurate trip estimates. Regarding operation time, the system can estimate based on task type and historical average processing efficiency, making cost calculations more closely reflect on-site conditions. For spare parts, the system can identify the distance and delivery patterns between different warehouses and adjust waiting times based on real-time supply chain updates. In actual task management, when a task requires cross-regional execution, the system can automatically add accommodation and allowance costs and automatically calculate the number of travel days based on the task's scope, making cost estimations more comprehensive.

[0036] In some embodiments, the total maintenance cost corresponding to the maintenance cost calculation data is calculated according to the following formula:

[0037] in, Let be the travel distance of the i-th maintenance task; Let be the number of business trip days corresponding to the i-th maintenance task; Let be the on-site operation time for the i-th maintenance task; Let i be the equipment downtime corresponding to the i-th maintenance task; The waiting time for the i-th maintenance task caused by the spare parts delivery time; This represents the probability of the i-th device failing within a preset prediction time range, based on the device status prediction model. , , , , , These are the unit price parameters for transportation costs, accommodation costs, catering costs, subsidies, labor costs, and equipment downtime losses, respectively. This is a parameter sensitive to downtime risk; This is a parameter sensitive to waiting time.

[0038] In practical implementation, the system can automatically select different parameter configurations based on different regions, task types, or equipment levels, enabling the cost calculation process to adapt to various operational and maintenance needs. In some cases, for high-risk equipment or core equipment on critical production lines, the system will automatically increase the risk sensitivity factor based on historical accident records, giving higher weight to downtime losses in cost evaluation. In scenarios with long material waiting times, the system will also automatically update the waiting ratio based on delivery progress, allowing related costs to be dynamically adjusted according to changes in the supply chain status. This approach makes the cost results closer to the actual situation on-site.

[0039] In some embodiments, inputting the prediction results and calculation results into the operation and maintenance decision model, and solving the operation and maintenance decision model includes: Task urgency parameters are constructed based on the prediction results of the future operating status of the equipment, and cost constraint parameters are constructed based on the operation and maintenance cost calculation data. The task urgency parameters and cost constraint parameters are used as input elements of the operation and maintenance decision model. Based on the availability information of maintenance personnel, the order of task execution, and the accessibility information of travel paths, personnel allocation constraints, task sequencing constraints, and path planning constraints are constructed. Input elements and constraints are input into the operation and maintenance decision model, and iterative solutions are performed on the operation and maintenance decision model to obtain a set of feasible solutions that satisfy the constraints. Select the target decision solution from the set of feasible solutions, which includes personnel allocation information, task sequence information, travel path information, and spare parts delivery information, as the operation and maintenance decision data.

[0040] During implementation, the system can automatically classify and merge tasks, clustering and filtering tasks with similar locations, equipment types, or maintenance requirements to reduce the solution scale and improve decision-making efficiency. During the solution process, the system can match tasks based on personnel skill levels, preventing personnel from being assigned to equipment types outside their skill range and ensuring the feasibility of the solution. For multi-region tasks, the system uses path filtering rules to avoid redundant round trips and prioritizes maintenance tasks for high-risk equipment based on equipment urgency, thus balancing safety and efficiency in the solution generation process.

[0041] In some embodiments, it also includes: During the execution of operation and maintenance tasks, real-time data is acquired to characterize the personnel execution status, spare parts delivery status, and equipment operation status. Executed or determined scheduling operations are set as fixed constraints, and the unexecuted scheduling parts are adjusted and calculated based on the updated operation and maintenance cost data expression to obtain updated operation and maintenance decision data.

[0042] In practice, the system can periodically collect data on personnel location changes, task completion progress, and equipment status changes, and flexibly update the original scheduling plan based on real-time data. When personnel are delayed or tasks are completed ahead of schedule, the system can automatically recalculate the execution order of the remaining tasks and adjust related costs to keep the scheduling plan optimal. In scenarios where spare parts are delayed or delivered early, the system will automatically update spare parts availability and adjust the execution window of subsequent tasks accordingly, allowing task scheduling to adapt to changes in the supply chain.

[0043] In some embodiments, adjusting the unexecuted scheduling portion based on the updated operation and maintenance cost data expression includes: Based on real-time data on personnel execution status, the current location, current task progress, and available execution time window of the target personnel are updated to form real-time constraint information that characterizes personnel availability. Based on real-time data on spare parts delivery status, the location of spare parts en route, estimated arrival time, and possible delay time are updated to form real-time constraint information to characterize material availability. Based on real-time data on equipment operating status, the operating trend, risk level changes, and allowable downtime window of the target equipment are updated to form real-time constraint information that characterizes the equipment demand status. The real-time constraints on personnel availability, material availability, and equipment demand, along with the updated operation and maintenance cost data expression, are input into the operation and maintenance decision model. The unexecuted scheduling parts are then re-solved to obtain a feasible scheduling solution that meets the real-time constraints. Select the scheduling result that is compatible with the original operation and maintenance decision data from the feasible scheduling solutions, and replace the corresponding unexecuted part in the original operation and maintenance decision data with the scheduling result to form the updated operation and maintenance decision data.

[0044] In practice, the system locks the results of completed tasks to ensure that resolving the problem does not affect the stability of the already executed parts. During the resolving process, the system can automatically filter out unreachable paths based on real-time changes in personnel locations and automatically adjust the task order based on changes in equipment risk, making the solution more aligned with actual on-site needs. For tasks executed collaboratively by multiple personnel, the system can automatically identify the temporal relationships within the team, allowing the updated schedule to be replaced while maintaining the overall structure, thus achieving a smooth update of the solution.

[0045] In some embodiments, this application provides an equipment operation and maintenance decision system for implementing an equipment operation and maintenance decision method, the system comprising: The data acquisition and preprocessing unit is used to acquire equipment operation data, maintenance personnel status data, spare parts resource data, and location service data, and to perform formatting and data alignment processing on the data to obtain a structured data set. The equipment status prediction unit is used to perform equipment status prediction processing based on equipment operation data in a structured dataset, and output prediction results to characterize the future operating status of the equipment. The operation and maintenance cost calculation unit is used to calculate the travel distance information, business trip days information, operation duration information and equipment downtime information of each operation and maintenance task based on location service data, status data and spare parts resource data, and generate operation and maintenance cost calculation data and total operation and maintenance cost according to cost parameter configuration. The operation and maintenance decision solving unit is used to input the prediction results output by the equipment status prediction unit and the operation and maintenance cost calculation data output by the operation and maintenance cost calculation unit into the operation and maintenance decision model, and solve it under personnel allocation constraints, task sequencing constraints and path planning constraints to obtain operation and maintenance decision data containing personnel allocation information, task sequence information, travel path information and spare parts delivery information. The dynamic adjustment unit for operation and maintenance is used to acquire real-time data on personnel execution status, spare parts delivery status, and equipment operation status during the execution of operation and maintenance tasks. It sets executed or determined scheduling operations as fixed constraints, and adjusts and calculates the unexecuted scheduling parts based on the updated operation and maintenance cost data to obtain updated operation and maintenance decision data.

[0046] In one specific embodiment, please refer to Figure 3 As shown, the equipment operation and maintenance decision-making system can be deployed according to the illustrated architecture. The system includes a data acquisition layer, a data processing layer, an optimization decision-making layer, an execution monitoring layer, and a feedback optimization layer, which logically coordinate with each functional unit. The data acquisition and preprocessing unit can connect to the equipment monitoring platform, personnel positioning terminals, spare parts management system, and map service interface to continuously receive equipment status data, personnel location data, spare parts inventory information, and travel resource information. The system can perform format conversion, time alignment, missing value imputation, and data cleaning on data from different sources, forming a structured data set that meets the needs of subsequent calculations.

[0047] The equipment status prediction unit can be used in conjunction with the fault prediction engine. It inputs the collected equipment operation sequence into the prediction model and generates predicted values ​​for future failure probabilities or remaining lifespan based on historical operating trends. The system can call the model updater during the prediction process to promptly replace the old model when new training parameters are provided by the feedback optimization layer, ensuring the prediction process remains sensitive to changes in the field. The cost calculation unit can be linked to the cost analyzer and resource status monitoring module. It comprehensively estimates operation and maintenance costs by using travel distance, operation duration, downtime, and spare parts waiting information, and combines this with parameters such as travel expenses, labor costs, and downtime losses to construct complete cost data for subsequent decision-making calculations.

[0048] During the scheduling process, the system utilizes a multi-objective optimizer to perform scheduling calculations, taking prediction results and cost data as input, and combining them with a constraint manager and a path planner to form a feasible solution space. The system filters feasible solutions based on current personnel availability, task priority relationships, accessibility to different areas, and spare parts delivery time windows. The scheduling decision-maker then generates a solution that meets the constraints. The generated solution includes information on personnel allocation, task sequence, route selection, and spare parts delivery, and is distributed to terminal devices for execution by a task dispatcher.

[0049] During the task execution phase, the system's execution monitoring layer can use cost monitors and execution trackers to record personnel travel progress, on-site execution status, and equipment operation changes in real time, and feed the relevant information back to the data processing layer. If the execution process is found to be inconsistent with the original scheduling plan, such as spare parts delays, personnel lag, or drastic changes in equipment status, the dynamic adjustment module can re-solve the unexecuted part based on real-time constraints to ensure the feasibility and coordination of the scheduling plan.

[0050] During the feedback optimization process, the system can aggregate execution results through an evaluation unit, including information such as scheduling cost deviations, task completion times, and risk events during execution. The strategy learner and parameter adjuster then adaptively optimize the prediction model, cost parameters, or scheduling strategies based on this feedback data to continuously improve the overall system decision-making quality. Through this hierarchical collaboration, the system achieves a closed-loop operation from data collection, status prediction, cost calculation, scheduling optimization to execution monitoring and strategy updates, effectively ensuring the real-time performance, accuracy, and resource utilization efficiency of the entire operation and maintenance process.

[0051] In some embodiments, this application provides a terminal, including: Memory, used to store device operation and maintenance decision-making programs; A processor is used to execute the steps of the equipment operation and maintenance decision-making system to implement the equipment operation and maintenance decision-making method.

[0052] In some embodiments, this application provides a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the device operation and maintenance decision-making method.

[0053] The above description is merely a preferred embodiment of one or more embodiments of this specification and is not intended to limit the scope of one or more embodiments of this specification. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments of this specification should be included within the protection scope of one or more embodiments of this specification.

Claims

1. A method for equipment operation and maintenance decision-making, characterized in that, Includes the following steps: Acquire initial data, which includes equipment operation data, maintenance personnel status data, spare parts resource data, and location service data. Preprocess the initial data to obtain a structured data set. Based on the equipment's operating data, perform equipment status prediction processing to obtain prediction results that reflect the future operating status of the equipment; Based on location service data, status data and spare parts resource data, the cost elements in the operation and maintenance process are quantified, and the cost elements are summarized one by one according to the preset cost accounting model to generate the corresponding operation and maintenance cost data. The prediction results and operation and maintenance cost data are input into the operation and maintenance decision model. Based on the objective function, constraints and optimization rules preset by the operation and maintenance decision model, the optimal or suboptimal solution that meets the conditions is searched in the solution space to obtain operation and maintenance decision data that includes personnel allocation information, task sequence information, travel path information and spare parts delivery information.

2. The equipment operation and maintenance decision-making method according to claim 1, characterized in that, The pre-trained equipment state prediction model performs equipment state prediction processing based on the equipment's operating data. The training of the equipment state prediction model includes the following steps: Acquire sample data containing historical operating parameters and historical fault records. Historical operating parameters are used to characterize the operating status of the equipment at different times, and historical fault records are used to indicate whether the equipment has experienced a fault or the time and location of the fault within the corresponding time period. Based on a preset time window, the historical operating parameters are divided into multiple consecutive operating state sequences in chronological order. Each operating state sequence is combined with its corresponding historical fault record to form a sample pair for training. A device status prediction model is established based on sample pairs, and the model is trained by comparing the correspondence between historical operating status and historical fault records. After the model training is completed, the sequence of the target device's operating state before the current moment is input into the device state prediction model to obtain the prediction results used to characterize the future operating state of the target device. The prediction results are the probability information of the device's failure and / or remaining operating life information within the preset prediction time range.

3. The equipment operation and maintenance decision-making method according to claim 1, characterized in that, The calculation of operation and maintenance costs based on location service data, status data, and spare parts resource data includes: Based on location service data, determine the travel distance information and corresponding travel time information between each maintenance personnel and each target device; Based on equipment status data and maintenance task plan data, determine the on-site operation time and equipment downtime information for each maintenance task; Based on spare parts resource data, determine the inventory status of spare parts required for each maintenance task, the spare parts arrival time, and the waiting time caused by the spare parts arrival time, and correct the corresponding equipment downtime information. The number of business trip days corresponding to each operation and maintenance task is determined based on calendar date information and start and end time information of operation and maintenance tasks. The unit price parameters for transportation costs, accommodation costs, catering costs, subsidy costs, labor costs, and equipment downtime losses are read from the cost parameter configuration. The travel distance information, business trip days information, on-site operation time information, and equipment downtime time information are correlated with the corresponding unit price parameters to form operation and maintenance cost calculation data that characterizes the transportation costs, accommodation costs, catering costs, subsidy costs, labor costs, and equipment downtime losses of each operation and maintenance task.

4. The equipment operation and maintenance decision-making method according to claim 3, characterized in that, The total maintenance cost corresponding to the maintenance cost calculation data is calculated according to the following formula: in, Let be the travel distance of the i-th maintenance task; Let be the number of business trip days corresponding to the i-th maintenance task; Let be the on-site operation time for the i-th maintenance task; Let i be the equipment downtime corresponding to the i-th maintenance task; The waiting time for the i-th maintenance task caused by the spare parts delivery time; This represents the probability of the i-th device failing within a preset prediction time range, based on the device status prediction model. , , , , , These are the unit price parameters for transportation costs, accommodation costs, catering costs, subsidies, labor costs, and equipment downtime losses, respectively. This is a parameter sensitive to downtime risk; This is a parameter sensitive to waiting time.

5. The equipment operation and maintenance decision-making method according to claim 1, characterized in that, The prediction and calculation results are input into the operation and maintenance decision model, and the solution to the operation and maintenance decision model includes: Task urgency parameters are constructed based on the prediction results of the future operating status of the equipment, and cost constraint parameters are constructed based on the operation and maintenance cost calculation data. The task urgency parameters and cost constraint parameters are used as input elements of the operation and maintenance decision model. Based on the availability information of maintenance personnel, the order of task execution, and the accessibility information of travel paths, personnel allocation constraints, task sequencing constraints, and path planning constraints are constructed. Input elements and constraints are input into the operation and maintenance decision model, and iterative solutions are performed on the operation and maintenance decision model to obtain a set of feasible solutions that satisfy the constraints. Select the target decision solution from the set of feasible solutions, which includes personnel allocation information, task sequence information, travel path information, and spare parts delivery information, as the operation and maintenance decision data.

6. The equipment operation and maintenance decision-making method according to any one of claims 1-5, characterized in that, Also includes: During the execution of operation and maintenance tasks, real-time data is acquired to characterize the personnel execution status, spare parts delivery status, and equipment operation status. Executed or determined scheduling operations are set as fixed constraints, and the unexecuted scheduling parts are adjusted and calculated based on the updated operation and maintenance cost data expression to obtain updated operation and maintenance decision data.

7. The equipment operation and maintenance decision-making method according to claim 6, characterized in that, The adjustment calculation for the unexecuted scheduling portion based on the updated operation and maintenance cost data expression includes: Based on real-time data on personnel execution status, the current location, current task progress, and available execution time window of the target personnel are updated to form real-time constraint information that characterizes personnel availability. Based on real-time data on spare parts delivery status, the location of spare parts en route, estimated arrival time, and possible delay time are updated to form real-time constraint information to characterize material availability. Based on real-time data on equipment operating status, the operating trend, risk level changes, and allowable downtime window of the target equipment are updated to form real-time constraint information that characterizes the equipment demand status. The real-time constraints on personnel availability, material availability, and equipment demand, along with the updated operation and maintenance cost data expression, are input into the operation and maintenance decision model. The unexecuted scheduling parts are then re-solved to obtain a feasible scheduling solution that meets the real-time constraints. Select the scheduling result that is compatible with the original operation and maintenance decision data from the feasible scheduling solutions, and replace the corresponding unexecuted part in the original operation and maintenance decision data with the scheduling result to form the updated operation and maintenance decision data.

8. An equipment operation and maintenance decision-making system, used to implement the equipment operation and maintenance decision-making method as described in claim 1, characterized in that, The system includes: The data acquisition and preprocessing unit is used to acquire equipment operation data, maintenance personnel status data, spare parts resource data, and location service data, and to perform formatting and data alignment processing on the data to obtain a structured data set. The equipment status prediction unit is used to perform equipment status prediction processing based on equipment operation data in a structured dataset, and output prediction results to characterize the future operating status of the equipment. The operation and maintenance cost calculation unit is used to calculate the travel distance information, business trip days information, operation duration information and equipment downtime information of each operation and maintenance task based on location service data, status data and spare parts resource data, and generate operation and maintenance cost calculation data and total operation and maintenance cost according to cost parameter configuration. The operation and maintenance decision solving unit is used to input the prediction results output by the equipment status prediction unit and the operation and maintenance cost calculation data output by the operation and maintenance cost calculation unit into the operation and maintenance decision model, and solve it under personnel allocation constraints, task sequencing constraints and path planning constraints to obtain operation and maintenance decision data containing personnel allocation information, task sequence information, travel path information and spare parts delivery information. The dynamic adjustment unit for operation and maintenance is used to acquire real-time data on personnel execution status, spare parts delivery status, and equipment operation status during the execution of operation and maintenance tasks. It sets executed or determined scheduling operations as fixed constraints, and adjusts and calculates the unexecuted scheduling parts based on the updated operation and maintenance cost data to obtain updated operation and maintenance decision data.

9. A terminal, characterized in that, include: Memory, used to store device operation and maintenance decision-making programs; A processor is used to implement the steps of the equipment operation and maintenance decision-making method as described in claim 1 when executing the equipment operation and maintenance decision-making device.

10. A computer-readable storage medium, characterized in that, The storage medium stores computer instructions. When the computer reads the computer instructions from the storage medium, the computer executes the equipment operation and maintenance decision-making method as described in claim 1.

Images