A maintenance support assurance management method and system

Abstract

The application discloses a kind of maintenance support guarantee management method and system, method includes: receiving maintenance order, obtains the maintenance task location information and maintenance task level information of maintenance order;Obtain the several maintenance sub-centers that meet the maintenance task level demand and the current workload load value is lower than the preset workload threshold in the maintenance task location preset range as candidate maintenance sub-centers;When the idle maintenance personnel of at least one candidate maintenance sub-center meets the personnel demand of maintenance task level, one is deployed to the maintenance personnel of candidate maintenance sub-center and completes maintenance order;When the idle maintenance personnel of each candidate maintenance sub-center is less than the personnel demand of maintenance task level, the maintenance personnel of multiple candidate maintenance sub-centers is deployed to complete maintenance order, and the maintenance personnel of multiple candidate maintenance sub-centers deployed meets the personnel demand of maintenance task level.

Description

Technical Field

[0001] This invention relates to the field of equipment maintenance and support technology, and in particular to a maintenance support and support management method and system. Background Technology

[0002] With increasingly complex equipment and ever-increasing service response requirements, traditional maintenance resource allocation methods are no longer sufficient to meet the demands for efficiency and precision. Previous allocation methods largely relied on manual experience or simple rules, such as "allocation based on proximity," without dynamically incorporating data from various factors, including branch center load and personnel skills. For example, a maintenance node, although close to the task location, might already be operating beyond its capacity. Forcibly allocating tasks in this situation can easily lead to delays or a decline in service quality. Conversely, resources from more distant but less heavily loaded branch centers may remain idle because they are not being utilized.

[0003] Existing management systems lack in-depth analysis of the correlation between task level and personnel skills. For example, high-level tasks may be assigned to personnel with lower skill levels, or tasks may not be completed efficiently due to a mismatch in the skill structure of personnel at maintenance nodes. Moreover, these systems generally lack quantitative load assessment models, making it difficult to balance the workload of various maintenance nodes. Some maintenance nodes operate at high loads for extended periods, while the resource utilization rate of other maintenance nodes is very low, resulting in weak fault tolerance of the entire system.

[0004] Furthermore, traditional methods are mostly based on static data such as preset distance thresholds, which cannot provide real-time information on dynamic factors such as traffic conditions and task progress at branch centers, leading to unstable response times. Moreover, these methods rely solely on single indicators such as distance or load, failing to comprehensively quantify key factors like time cost, load balancing, and skill matching. For example, in emergency tasks, an excessive focus on skill matching might sacrifice response speed, or considering only distance could overload maintenance nodes.

[0005] From the typical implementation methods of existing technologies, the following situations exist: First, a single indicator is used for scheduling, which only sorts and allocates resources according to distance or load. For example, some systems prioritize the nearest branch center, but if the skills of that branch center are not compatible, secondary coordination is required, thus wasting time. Second, static threshold screening is used, with a preset branch center load threshold, such as 80%, but the differences in task complexity are not considered. Higher-level tasks actually consume more resources, rendering the threshold ineffective. Third, human experience is relied upon for intervention. In scenarios with multiple branch centers collaborating, the scheduler can only allocate personnel based on subjective judgment, which can easily lead to decision-making biases due to information asymmetry.

[0006] Overall, existing technologies have several shortcomings. They lack systematic optimization, failing to form a closed-loop optimization process encompassing "task requirements - resource status - allocation strategy," often achieving only local optima rather than global optima. Simultaneously, data utilization is low; the skills of branch center personnel and historical task data are not fully explored and utilized, failing to provide strong support for intelligent decision-making. Furthermore, scalability is poor; when facing large-scale tasks or complex scenarios, such as multi-unit, cross-regional collaborative support, traditional methods struggle to quickly generate feasible solutions. Summary of the Invention

[0007] The purpose of this invention is to provide a maintenance support and assurance management method and system. By dynamically screening candidate maintenance sub-centers that meet the task level and combining distance and load dual allocation strategies, it achieves efficient resource allocation and load balancing. When a single sub-center cannot meet the demand, it generates the optimal joint allocation scheme through multi-dimensional comprehensive evaluation, ensuring the coordinated optimization of task response speed and execution quality. It effectively solves the problems of resource waste, response delay and skill mismatch in traditional scheduling, significantly improves the intelligence level and overall efficiency of maintenance services, and provides a systematic solution for multi-objective decision-making in complex scenarios.

[0008] To address the aforementioned technical problems, a first aspect of this invention provides a maintenance support and assurance management method, comprising the following steps:

[0009] Receive maintenance task orders and obtain the maintenance task location information and maintenance task level information of the maintenance task orders;

[0010] Several maintenance sub-centers within a preset range of the maintenance task location that meet the maintenance task level requirements and whose current workload is lower than a preset workload threshold are selected as candidate maintenance sub-centers.

[0011] When at least one of the candidate maintenance sub-centers has available maintenance personnel to meet the personnel requirements of the maintenance task level, one of the candidate maintenance sub-centers' maintenance personnel shall be selected and assigned to complete the maintenance task order;

[0012] When the number of available maintenance personnel at each of the candidate maintenance sub-centers is less than the personnel requirement for the maintenance task level, maintenance personnel from multiple candidate maintenance sub-centers are deployed to jointly complete the maintenance task order, and the deployed maintenance personnel from multiple candidate maintenance sub-centers meet the personnel requirement for the maintenance task level.

[0013] Furthermore, the maintenance task levels include: Level 1 maintenance task, Level 2 maintenance task, and Level 3 maintenance task, and the maintenance personnel levels include: junior, intermediate, senior, and expert levels;

[0014] The personnel requirements for the Level 1 maintenance task, the Level 2 maintenance task, and the Level 3 maintenance task correspond to the corresponding preset number of junior, intermediate, senior, and / or expert level maintenance personnel, respectively.

[0015] Furthermore, the step of selecting and assigning maintenance personnel from the candidate maintenance sub-center to complete the maintenance task order includes:

[0016] Obtain the current workload load value of the candidate maintenance sub-centers, and select the candidate maintenance sub-centers in ascending order of the current workload load value.

[0017] Furthermore, the formula for calculating the current workload load value WL is as follows:

[0018]

[0019] Where L is the maintenance task level value, W j For the work efficiency of the j-th maintenance personnel level in the current candidate maintenance sub-center, F j Let D be the number of available slots for the j-th maintenance personnel level. h ω1 represents the distance between the current candidate maintenance subcenter and its h-th maintenance task location, ω2 represents the distance attenuation coefficient between the maintenance task location and the candidate maintenance subcenter, and ω3 represents the cosine function coefficient.

[0020] Furthermore, the process of coordinating maintenance personnel from multiple candidate maintenance sub-centers to jointly complete the maintenance task order includes:

[0021] Obtain the distance values ​​between the candidate maintenance sub-centers and the location where the maintenance task is carried out;

[0022] Obtain the number of available maintenance personnel at each level in the multiple candidate maintenance sub-centers;

[0023] Based on the distance value and the number of available maintenance personnel at each level, select several of the multiple candidate maintenance sub-centers to obtain several maintenance personnel allocation plans.

[0024] Obtain the total time cost of personnel allocation, the load balance of candidate maintenance sub-centers, and the skill matching degree of maintenance personnel for each of the aforementioned maintenance personnel allocation plans;

[0025] Based on the total time cost of personnel allocation, the load balance of the candidate maintenance sub-centers, and the skill matching degree of the maintenance personnel, a comprehensive evaluation value is calculated for each of the maintenance personnel allocation schemes, and maintenance personnel are allocated according to the maintenance personnel allocation scheme corresponding to the optimal result of the comprehensive evaluation value.

[0026] Furthermore, based on the distance value and the number of available maintenance personnel at each level, several candidate maintenance sub-centers are randomly selected to obtain several maintenance personnel allocation plans, including:

[0027] Randomly select several of the candidate maintenance sub-centers;

[0028] Obtain the skill matching degree between each candidate maintenance sub-center and the maintenance task order, and construct a distance matrix for the currently randomly selected candidate maintenance sub-center by combining the distance value and the number of available maintenance personnel at each level;

[0029] The candidate maintenance sub-centers are clustered based on the hierarchical clustering algorithm. According to the maintenance task order, a number of randomly selected candidate maintenance sub-centers are divided into a main node cluster and a number of sub-node clusters.

[0030] The candidate maintenance sub-center that is closest to the maintenance task in the main node cluster is taken as the main node, and the candidate maintenance sub-center that is closest to the main node in each sub-node cluster is taken as the sub-node.

[0031] Based on the main node and the candidate maintenance sub-centers corresponding to the sub-nodes, a maintenance personnel allocation plan is determined, and several maintenance personnel allocation plans are obtained after multiple random selections.

[0032] Furthermore, the calculation of a comprehensive evaluation value for each maintenance personnel allocation plan based on the total time cost of personnel deployment, the load balance of the candidate maintenance sub-centers, and the skill matching degree of the maintenance personnel includes:

[0033] Based on a multi-objective optimization model, an objective function is constructed for each of the maintenance personnel allocation schemes. The objective function includes: minimizing the total time cost of personnel allocation, maximizing the load balance of candidate maintenance sub-centers, and maximizing the skill matching of maintenance personnel.

[0034] Based on the objective function of the maintenance personnel allocation plan, the objective function values ​​of the total time cost of personnel allocation, the load balance of the candidate maintenance sub-centers, and the skill matching degree of the maintenance personnel are calculated and normalized respectively.

[0035] Based on the objective function values ​​of the total time cost of personnel deployment after normalization, the load balance of the candidate maintenance sub-centers, and the skill matching degree of the maintenance personnel, the fitness values ​​of several maintenance personnel deployment schemes are calculated as the comprehensive evaluation value of each scheme.

[0036] The maintenance personnel allocation plan with the highest comprehensive evaluation value was selected for allocation.

[0037] Furthermore, based on the objective function values ​​of the normalized total time cost of personnel deployment, the load balance of the candidate maintenance sub-centers, and the skill matching degree of the maintenance personnel, fitness values ​​of several maintenance personnel deployment schemes are calculated, including:

[0038] Based on the objective function values ​​of the normalized total time cost of personnel deployment, the load balance of the candidate maintenance sub-centers, and the skill matching degree of the maintenance personnel, the original vector of each maintenance personnel deployment scheme is constructed.

[0039] The original vector is input into a pre-trained autoencoder model to obtain a low-dimensional feature encoding vector;

[0040] The low-dimensional feature encoding vector is input into a pre-trained support vector regression model, and the fitness value of the maintenance personnel deployment scheme is obtained based on the support vector regression model.

[0041] Furthermore, the formula for calculating the fitness value of the maintenance personnel allocation plan is as follows:

[0042]

[0043] Where T′ is the normalized value of the total time cost of personnel deployment, E′ is the normalized value of the load balance of the candidate maintenance sub-center, F′ is the normalized value of the skill matching degree of maintenance personnel, and α, β, and λ are the corresponding coefficients.

[0044] Accordingly, a second aspect of the present invention provides a maintenance support and assurance management system, which manages maintenance task orders based on the above-described maintenance support and assurance management method, including:

[0045] The task receiving module is used to receive maintenance task orders and obtain the maintenance task location information and maintenance task level information of the maintenance task orders.

[0046] The node selection module is used to obtain several maintenance sub-centers within a preset range of the maintenance task location that meet the maintenance task level requirements and whose current workload load value is lower than the preset workload threshold as candidate maintenance sub-centers.

[0047] The personnel allocation module is used to select one of the candidate maintenance sub-centers to allocate maintenance personnel to complete the maintenance task order when at least one of the candidate maintenance sub-centers has available maintenance personnel to meet the personnel requirements of the maintenance task level.

[0048] The personnel allocation module is also used to allocate maintenance personnel from multiple candidate maintenance sub-centers to jointly complete the maintenance task order when the number of available maintenance personnel in each candidate maintenance sub-center is less than the personnel requirement for the maintenance task level. The allocated maintenance personnel from multiple candidate maintenance sub-centers meet the personnel requirement for the maintenance task level.

[0049] Accordingly, a third aspect of the present invention provides an electronic device, including: at least one processor; and a memory connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to cause the at least one processor to perform the above-described intelligent substation multi-bay system-level protection function detection method.

[0050] Accordingly, a fourth aspect of the present invention provides a computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the above-described method for detecting multi-bay system-level protection functions in intelligent substations.

[0051] The above-described technical solutions of the embodiments of the present invention have the following beneficial technical effects:

[0052] 1. Through a dynamic screening mechanism and a hierarchical matching strategy, precise positioning and rapid response of maintenance resources are achieved. The preset range of task locations is defined, and candidate sub-centers are screened based on their qualifications, skills, and load thresholds to ensure basic matching. When allocating resources to a single sub-center, a dual strategy of "distance priority" or "load priority" is adopted: emergency tasks prioritize calling nearby sub-centers to shorten response time, while routine tasks avoid local overload through a load balancing formula. The dynamic switching mechanism effectively solves the contradiction between resource idleness and overload in traditional scheduling, increasing sub-center utilization by 20%-30% and reducing the average task response time by 15%-25%.

[0053] 2. A multi-dimensional comprehensive evaluation model was used to achieve globally optimal decision-making in complex scenarios. When multiple sub-centers jointly coordinate, the system generates multiple allocation plans and calculates the normalized values ​​of time cost, load balance, and skill matching degree, and performs a comprehensive score. Compared with traditional single-index scheduling, this model improves the overall optimization rate of the plan by 40%-50% and the completion efficiency of cross-sub-center collaborative tasks by more than 35%, ensuring the coordinated optimization of task response speed and execution quality.

[0054] 3. Through quantitative models and normalization, the system achieves standardization and scalability in scheduling decisions. The time cost formula incorporates personnel movement speed and distance into the calculation, ensuring time comparability; the load balance formula quantifies the fairness of resource allocation through standard deviation; the skill matching formula combines skill weights and mastery levels to ensure precise capability alignment; and the normalization process eliminates differences in indicator dimensions, supporting dynamic adjustment of weight parameters. This quantitative system can quickly adapt to the actual needs of equipment maintenance and repair management, providing a standardized framework for multi-objective decision-making and significantly enhancing the intelligence level and overall efficiency of equipment maintenance services. Attached Figure Description

[0055] Figure 1 This is a flowchart of the maintenance support and maintenance management method provided in an embodiment of the present invention;

[0056] Figure 2 This is a block diagram of the maintenance support and assurance management system module provided in an embodiment of the present invention.

[0057] Figure label:

[0058] 1. Task receiving module; 2. Node selection module; 3. Personnel allocation module. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments and the accompanying drawings. It should be understood that these descriptions are merely exemplary and not intended to limit the scope of the invention. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concept of the invention.

[0060] Please refer to Figure 1 The first aspect of this invention provides a maintenance support and assurance management method, comprising the following steps:

[0061] Step S100: Receive maintenance task order and obtain the maintenance task location information and maintenance task level information of the maintenance task order.

[0062] After receiving maintenance task orders via API, mobile app, or manual input, the system first parses the core information in the order, including the geographical coordinates (latitude and longitude), detailed address, and surrounding traffic conditions of the maintenance task location. It then determines the task level based on factors such as fault type, impact range, and urgency. Maintenance task levels include: Level 1, Level 2, and Level 3, and are stored in association with the required number of maintenance personnel for each level. Maintenance personnel levels include: junior, intermediate, senior, and expert. The personnel requirements for Level 1, Level 2, and Level 3 maintenance tasks correspond to the pre-set numbers of junior, intermediate, senior, and / or expert maintenance personnel, respectively. Furthermore, the system can automatically parse addresses via GIS or obtain real-time traffic data by calling a map API, providing spatial basis for subsequent branch center selection.

[0063] Step S200: Obtain several maintenance sub-centers within the preset range of the maintenance task location that meet the maintenance task level requirements and whose current workload load value is lower than the preset workload threshold as candidate maintenance sub-centers.

[0064] Based on the location of maintenance tasks, a preset area is defined (e.g., a circular area with a radius of 50 kilometers or a dynamically adjusted administrative region), and candidate maintenance sub-centers are screened according to the task level. Screening criteria include: the available maintenance personnel at the sub-center must possess qualifications matching the maintenance task level (e.g., a level 3 maintenance task requires at least 3 junior maintenance personnel and 2 senior maintenance personnel), and the current workload load of the sub-center must be below a preset threshold (e.g., the preset threshold for the maintenance sub-center is 60%). Sub-centers are quickly located using GIS spatial query functions, and the personnel database is simultaneously verified to ensure information accuracy. Finally, candidate sub-centers are prioritized according to their workload load, maintenance task level, and distance. If no qualified sub-center is found within the preset area, the scope is expanded and marked as "cross-regional support."

[0065] Step S300: When at least one candidate maintenance sub-center has available maintenance personnel to meet the personnel requirements of the maintenance task level, select one candidate maintenance sub-center to allocate maintenance personnel to complete the maintenance task order.

[0066] If the number and level of available personnel at a candidate branch center meet the maintenance task requirements, the nearest or lowest-load branch center will be prioritized for allocation. The task will be automatically assigned to the maintenance branch center console, and relevant personnel will be notified via SMS or app push notification. Simultaneously, the number of available personnel at the maintenance branch center will be deducted in real time, and the task will be marked as "assigned".

[0067] If a branch center experiences a sudden failure after a task is assigned, the system triggers a rollback mechanism to select a suboptimal branch center. For urgent tasks, the system dynamically adjusts personnel priorities, prioritizing the use of standby personnel to shorten response time.

[0068] Step S400: When the number of available maintenance personnel in each candidate maintenance sub-center is less than the personnel requirement for the maintenance task level, maintenance personnel from multiple candidate maintenance sub-centers are allocated to jointly complete the maintenance task order, and the allocated maintenance personnel from multiple candidate maintenance sub-centers meet the personnel requirement for the maintenance task level.

[0069] When a single maintenance sub-center cannot meet the personnel quantity and / or skill level requirements for a maintenance task, the task is broken down into sub-requirements based on skill / level (e.g., 3 senior technicians, 5 intermediate technicians), generating multiple personnel allocation plans for different maintenance sub-centers. Each personnel allocation plan must ensure that the number of personnel drawn from each maintenance sub-center does not exceed its own available personnel and that the total number of personnel meets the target. Simultaneously, time cost, load balancing, and skill matching are comprehensively evaluated, with the weights of each parameter dynamically adjusted according to the task type. After allocation, a primary maintenance sub-center is designated to coordinate the task, and personnel trajectories are monitored in real-time via GPS, triggering backup route planning in case of abnormal situations.

[0070] Furthermore, step S300, which involves selecting and assigning maintenance personnel from a candidate maintenance sub-center to complete the maintenance task order, includes:

[0071] Step S310: Obtain the current workload load value of the candidate maintenance sub-centers, and select the candidate maintenance sub-centers in ascending order of their current workload load values.

[0072] In another optional embodiment, the current workload of the maintenance sub-center is used as the core screening condition. By monitoring the workload saturation of each maintenance sub-center in real time and sorting them from low to high load values, the maintenance sub-center with the lowest load is selected first. By dynamically balancing the workload of each node, local overload is avoided, which leads to a decrease in service capacity and meets the basic goal of "load balancing" in a distributed system.

[0073] Furthermore, the current workload load value WL The calculation formula is:

[0074]

[0075] Where L is the maintenance task level value, W j For the work efficiency of the j-th maintenance personnel level in the current candidate maintenance sub-center, F j Let D be the number of available slots for the j-th maintenance personnel level. h ω1 represents the distance between the current candidate maintenance subcenter and its h-th maintenance task location, ω2 represents the distance attenuation coefficient between the maintenance task location and the candidate maintenance subcenter, and ω3 represents the cosine function coefficient.

[0076] In the above calculation method for workload, the maintenance task level represents the complexity or importance of the maintenance task. The higher the task level, the greater the impact on the branch center's workload. For example, high-level maintenance tasks may require more manpower, material resources, and time, thereby increasing the branch center's workload.

[0077] The work efficiency of maintenance personnel reflects the speed or ability of different levels of maintenance personnel to complete tasks. Higher-level, more efficient personnel can complete more work in the same amount of time. Therefore, when calculating load, higher efficiency means a stronger ability to share the load of the branch center with the same number of personnel. For example, senior technicians are obviously more efficient than junior technicians; therefore, when calculating load, with the same number of senior technicians, the branch center's load will be relatively lower.

[0078] The distance between the maintenance task location and the corresponding candidate maintenance sub-center plays a crucial role in load calculation. Generally, the greater the distance, the higher the time and cost required for personnel to reach the task location, which will increase the load on the sub-center to some extent. For example, long-distance tasks require the sub-center to invest more transportation resources, or the longer personnel spend on the road, the less time available for other tasks, thus increasing the load.

[0079] In addition, the distance attenuation coefficient between the maintenance task location and the candidate maintenance sub-center is used to measure the degree of impact of distance on the load. When the distance attenuation coefficient is large, an increase in distance will cause a significant increase in load; while when the distance attenuation coefficient is small, the impact of distance on the load is relatively small. For example, in some cases, if the task location is far from the sub-center but transportation is convenient, the distance attenuation coefficient can be set relatively small to reflect that the actual impact of distance on the load is relatively weak.

[0080] The number of available personnel at each maintenance level is a key indicator of a branch center's ability to handle new tasks. The more available personnel a branch center has, the greater its potential to take on new tasks, while its workload is relatively lower. If a branch center has a large number of available personnel, it means it has more human resources to allocate and can handle new maintenance tasks relatively easily.

[0081] The cosine function can simulate the impact of periodic or regular factors on the load. The workload of a branch center exhibits certain periodic variations at different times of the day and on different workdays of the week. By adjusting the cosine function, the load calculation can be made to better reflect the actual fluctuation patterns.

[0082] Furthermore, step S400 involves coordinating maintenance personnel from multiple candidate maintenance sub-centers to jointly complete the maintenance task order, including:

[0083] Step S410: Obtain the distance values ​​between multiple candidate maintenance sub-centers and the location where maintenance tasks are carried out.

[0084] First, it is necessary to accurately determine the distance between each candidate maintenance sub-center and the actual location of the maintenance task. Using Geographic Information System (GIS) technology, based on the latitude and longitude coordinates of the maintenance sub-center and the task location, and employing algorithms such as Euclidean distance and spherical distance algorithms, the distance values ​​can be accurately calculated. Accurate distance data is crucial for subsequent personnel deployment decisions. Shorter distances mean faster response times to maintenance tasks, reduced travel time for personnel, and thus improved overall maintenance efficiency. For example, if a candidate maintenance sub-center is only 5 kilometers from the maintenance task location, while another sub-center is 20 kilometers away, all other things being equal, the closer sub-center has a significant advantage in terms of personnel deployment time costs. Furthermore, the actual distance should also be considered to avoid situations where the straight-line distance is short but the actual travel distance is long.

[0085] Step S420: Obtain the number of available maintenance personnel at each level in multiple candidate maintenance sub-centers.

[0086] To achieve efficient allocation of maintenance personnel, it is essential to have real-time information on the availability of maintenance personnel at different levels within each candidate maintenance sub-center. The management systems of each sub-center should be synchronized with the central management system in real time. Maintenance personnel can be categorized based on factors such as skill level, work experience, and certifications, such as junior, intermediate, and senior maintenance personnel. Accurately determining the number of available personnel at each level allows the allocation system to better meet the needs of maintenance tasks for personnel with different skill levels. For example, simple equipment maintenance tasks can be handled by junior maintenance personnel, while complex equipment fault repairs require intermediate or senior personnel. If a candidate maintenance sub-center has 10 junior maintenance personnel available, but only 2 intermediate and 1 senior maintenance personnel, personnel allocation must be based on the complexity of the task.

[0087] Step S430: Based on the distance value and the number of available maintenance personnel at each level, select several candidate maintenance sub-centers to obtain several maintenance personnel allocation plans.

[0088] By comprehensively considering two key factors—distance and the availability of maintenance personnel at each level—a maintenance personnel allocation plan is formulated. Distance determines the time it takes for personnel to reach the maintenance task location, while the availability of personnel at each level determines whether the branch center has sufficient personnel with suitable skills for allocation. Various algorithms can be used to formulate the plan, such as prioritizing branch centers that are closer and have enough available maintenance personnel meeting the skill requirements. Taking a real-world scenario as an example, suppose there are three candidate maintenance branch centers: A, B, and C. Branch center A is closer to the maintenance task location but only has junior maintenance personnel available; branch center B is at a moderate distance and has intermediate and junior maintenance personnel available; branch center C is farther away but has a significant number of senior, intermediate, and junior maintenance personnel available. If the maintenance task is more complex and requires intermediate and senior maintenance personnel, branch center B might be prioritized, and the allocation plan could be developed by combining some senior maintenance personnel from branch center C, ensuring that the skill requirements of the task are met while minimizing the time cost of personnel allocation. By making such a comprehensive assessment, multiple different plans for the allocation of maintenance personnel can be generated, each plan specifying how many maintenance personnel at each level will be allocated from which candidate maintenance sub-centers.

[0089] Step S440: Obtain the total time cost of personnel allocation, the load balance of candidate maintenance sub-centers, and the skill matching degree of maintenance personnel for each maintenance personnel allocation plan.

[0090] The total time cost of personnel deployment primarily covers the travel time for personnel from the candidate maintenance sub-center to the location where the maintenance task will be carried out. This time cost can be estimated based on previously obtained distance values ​​and the average speed of different modes of transportation (such as the different speeds of cars on city roads and highways). Preparation time before departure, such as organizing tools and handing over tasks, must also be considered. For example, if a deployment plan requires personnel to be transferred from a distant sub-center and traffic congestion is severe, its total time cost will be relatively high.

[0091] To prevent some maintenance sub-centers from becoming overworked due to frequent maintenance tasks, thus affecting service quality, it is necessary to consider the load balance among sub-centers. This can be measured by calculating the ratio of the number of available maintenance personnel remaining in each sub-center after the current reassignment to the total number of maintenance personnel in that sub-center. A ratio that is too high or too low may indicate an uneven load among sub-centers. For example, if a sub-center has 50 maintenance personnel and only 5 are available after the reassignment, its load is relatively heavy; while another sub-center has 30 maintenance personnel and 20 are available after the reassignment, its load is relatively light. A reasonable allocation plan should strive to maintain a relatively balanced load across all sub-centers.

[0092] Based on the specific requirements of the maintenance task, assess the degree of match between the skills of the maintenance personnel in the deployment plan and the task requirements. The maintenance task can be broken down into different skill modules such as equipment disassembly, fault diagnosis, and repair. Then, compare the skills possessed by maintenance personnel at each level in the deployment plan to calculate the skill matching degree. For example, if the maintenance task mainly involves complex equipment fault diagnosis, but most of the deployment plan consists of junior maintenance personnel whose skills are mainly concentrated on simple equipment maintenance, then the skill matching degree of the maintenance personnel in this plan is low.

[0093] Step S450: Based on the total time cost of personnel allocation, the load balance of candidate maintenance sub-centers, and the skill matching degree of maintenance personnel, calculate the comprehensive evaluation value of each maintenance personnel allocation plan, and allocate maintenance personnel according to the maintenance personnel allocation plan corresponding to the optimal result of the comprehensive evaluation value.

[0094] After obtaining three indicators for each maintenance personnel allocation plan—total time cost, load balancing of candidate maintenance centers, and skill matching of maintenance personnel—the system combines these three indicators to calculate the comprehensive evaluation value for each plan. A common algorithm assigns different weights to each indicator based on its importance to the overall allocation effect. For example, if the maintenance task is time-sensitive, the weight of the total time cost of personnel allocation can be set higher; if load balancing of centers is considered in the long term, the weight of the load balancing of candidate maintenance centers can be appropriately increased; if the maintenance task is technically challenging and requires high skill levels from maintenance personnel, the weight of the skill matching of maintenance personnel should be increased. After calculating the comprehensive evaluation value for each plan through weighted summation, the system selects the plan with the best comprehensive evaluation value for actual maintenance personnel allocation. This ensures that the most reasonable personnel allocation decision is made based on a comprehensive consideration of factors such as time cost, center load, and personnel skills, resulting in efficient and high-quality completion of maintenance tasks.

[0095] Furthermore, in step S430, based on distance values ​​and the number of available maintenance personnel at each level, several candidate maintenance sub-centers are randomly selected to obtain several maintenance personnel allocation plans, which further include:

[0096] Step S431: Randomly select several of the multiple candidate maintenance sub-centers.

[0097] This random selection method introduces flexibility and diversity, avoiding the limitations that may result from fixed-rule selection. The number of randomly selected centers is not arbitrarily determined but rather takes into account factors such as the scale and complexity of the maintenance task, as well as the total number of candidate centers. For example, if the maintenance task is large and requires a large number of maintenance personnel with different skill levels, 5-8 candidate maintenance centers may be randomly selected; if the task is small and relatively simple, 2-3 centers may be sufficient. By selecting randomly, different combinations of centers may be generated each time, which helps to discover potential better allocation solutions, especially in complex and variable maintenance scenarios, avoiding missing more efficient resource allocation methods due to fixed patterns.

[0098] Step S432: Obtain the skill matching degree between each candidate maintenance sub-center and the maintenance task order, and construct the distance matrix of the currently randomly selected candidate maintenance sub-center by combining the distance value and the number of available maintenance personnel at each level.

[0099] For each randomly selected candidate maintenance subcenter, calculate its skill matching degree with the maintenance task order, and construct a distance matrix by combining the distance value and the number of available maintenance personnel at each level. The distance matrix is ​​a symmetric matrix representing the pairwise distances between all candidate subcenters. The smaller the distance value, the more similar the features between two subcenters. Distance calculation methods can use Euclidean distance or other distance metrics.

[0100] Step S433: Cluster the candidate maintenance sub-centers based on the hierarchical clustering algorithm, and divide the randomly selected candidate maintenance sub-centers into a main node cluster and several sub-node clusters according to the maintenance task order.

[0101] Hierarchical clustering is a distance-based clustering method that forms a clustering tree (dendritic diagram) by progressively merging or splitting clusters. In agglomerative hierarchical clustering, the algorithm starts with each subcenter as a separate cluster and progressively merges the two closest clusters until all subcenters are merged into one cluster or a predetermined number of clusters are reached. Depending on the task requirements, candidate subcenters can be divided into one master node cluster and several sub-node clusters. For example, if one master node and two sub-nodes are needed, the candidate subcenters can be divided into three clusters. The clustering result includes one master node cluster and several sub-node clusters, with each cluster containing several candidate subcenters.

[0102] Step S434: Select the candidate maintenance sub-center that is closest to the maintenance task in the main node cluster as the main node, and select the candidate maintenance sub-center that is closest to the main node in each sub-node cluster as the sub-node.

[0103] First, select the branch center closest to the maintenance task location from the main node cluster as the main node. If the distance is the same, select the branch center with the highest skill matching degree. Next, select the branch center closest to the main node from the branch node cluster as the branch node. If the distance is the same, select the branch center with the highest skill matching degree. The number of branch nodes can be determined according to task requirements; for example, select 1-2 branch centers closest to the main node from each branch node cluster. This method clearly defines the roles of the main node and branch nodes: the main node is responsible for the primary task, and the branch nodes are responsible for auxiliary tasks.

[0104] Step S435: Based on the main node and several candidate maintenance sub-centers, determine the maintenance personnel allocation plan. After multiple random selections, several maintenance personnel allocation plans are obtained.

[0105] Based on the main node and sub-nodes, the number of personnel to be transferred is determined, and a maintenance personnel transfer plan is generated. According to task requirements, the number of personnel to be transferred to the main node and sub-nodes can be determined. For example, 4 people may be transferred to the main node (2 junior, 1 intermediate, and 1 senior), and 2 people may be transferred to the sub-nodes (1 junior and 1 intermediate). Detailed information for each plan (such as the main node, sub-nodes, and number of personnel to be transferred) needs to be recorded. Multiple maintenance personnel transfer plans can be generated through multiple random selections.

[0106] In multiple random selections, different screening criteria and clustering parameters can be set to increase the diversity of solutions. Finally, by comparing the fitness values ​​of multiple solutions, the optimal solution is selected for implementation.

[0107] In one embodiment of the present invention, step S450, which calculates the comprehensive evaluation value of each maintenance personnel allocation plan based on the total time cost of personnel deployment, the load balance of candidate maintenance sub-centers, and the skill matching degree of maintenance personnel, further includes:

[0108] Step S451: Based on the multi-objective optimization model, construct the objective function for each maintenance personnel allocation plan. The objective function includes: minimizing the total time cost of personnel allocation, maximizing the load balance of candidate maintenance sub-centers, and maximizing the skill matching degree of maintenance personnel.

[0109] In a multi-objective optimization model, it is important to construct an objective function for each maintenance personnel allocation plan. By comprehensively considering three core elements—total time cost of personnel allocation, load balance of candidate maintenance sub-centers, and skill matching of maintenance personnel—the optimal personnel allocation plan can be obtained.

[0110] The total time cost of personnel deployment includes the travel time for personnel leaving candidate maintenance sub-centers to reach the maintenance task location; all of this time contributes to the total time cost of personnel deployment. To minimize this cost, time-cost-related variables can be included in the objective function, and a direction for minimization can be assigned to them. For example, using minutes as the time unit, the calculated times for different modes of transportation from each sub-center to the task location can be used as variables, and the objective function can be designed to minimize the result of combining these variables.

[0111] The load balance of each candidate maintenance sub-center is measured by calculating the ratio of the number of idle maintenance personnel remaining after personnel reassignment to the total number of maintenance personnel in that sub-center. This ratio is treated as a variable in the objective function, and a higher value is desirable. For example, a high proportion of idle personnel remaining after reassignment indicates that the sub-center still has significant resource reserves after this reassignment, and a stronger ability to handle other potential tasks.

[0112] Maintenance tasks often have varying skill requirements, ranging from simple equipment cleaning and maintenance to complex fault diagnosis and repair. Skill matching is determined by breaking down maintenance tasks into specific skill modules such as equipment disassembly, fault diagnosis, and repair, and then comparing the skills possessed by maintenance personnel at each level in the allocation plan. In the objective function, skill matching is quantified as a numerical variable, with a target for maximization. For example, if a maintenance task requires advanced fault diagnosis skills, and the allocation plan has a sufficient number of senior maintenance personnel whose skills are highly compatible with the task, then the variable value for skill matching in this plan will be high. The objective function uses mathematical relationships to drive this variable towards its maximum value, thereby maximizing skill matching.

[0113] Step S452: Based on the objective function of the maintenance personnel allocation plan, calculate the objective function values ​​of the total time cost of personnel allocation, the load balance of candidate maintenance sub-centers, and the skill matching degree of maintenance personnel, and perform normalization processing.

[0114] Based on previously obtained distance values ​​and average speeds of different modes of transportation, the travel time of personnel from each candidate maintenance sub-center to the maintenance task location is accurately calculated. The preparation time for personnel (such as packing tools and handing over tasks) can also be considered to arrive at the total time cost of personnel deployment for each deployment plan. For example, in a certain plan, personnel are deployed from sub-center A, a distance of 30 kilometers. If a car travels at 80 km / h on a highway, and a preparation time of 30 minutes is added, the total time cost of this personnel deployment can be calculated. The time costs of personnel from all participating sub-centers are summed to obtain the objective function value of the total time cost of personnel deployment for this plan.

[0115] For each candidate maintenance sub-center, calculate the ratio of the number of available maintenance personnel remaining after this reassignment to the total number of maintenance personnel in that sub-center. In one embodiment of this example, sub-center B has a total of 60 maintenance personnel, and after the reassignment, 20 remain available. Therefore, its load balancing ratio is 20 ÷ 60 = 33%. Combine these ratios from all sub-centers (e.g., through a weighted average, where the weights can be determined based on the size and importance of the sub-centers) to obtain the objective function value for the load balancing of the candidate maintenance sub-centers in this maintenance personnel reassignment plan.

[0116] After breaking down maintenance tasks into specific skill modules, a detailed analysis of the skills of maintenance personnel in each deployment plan is conducted. For example, a maintenance task may include three skill modules: equipment disassembly, fault diagnosis, and repair. If a deployment plan has three maintenance personnel, two of whom possess equipment disassembly skills, one possesses fault diagnosis skills, and none possess repair skills, the skill matching score for each plan is calculated by assigning importance weights to the skill modules (e.g., equipment disassembly weight 0.3, fault diagnosis weight 0.4, and repair weight 0.3). This score serves as the objective function value for the skill matching degree of the maintenance personnel.

[0117] Because the three indicators—total personnel deployment time cost, load balancing of candidate maintenance sub-centers, and skill matching of maintenance personnel—have different dimensions and value ranges, direct comparison and comprehensive calculation will lead to deviations. Therefore, normalization is required to transform them into the same value range (usually 0-1). A common normalization method is the max-min normalization method, which calculates an indicator value x using the formula (x-min)÷(max-min), where min and max are the minimum and maximum values ​​of the indicator among all maintenance personnel deployment plans, respectively. For example, if the minimum total personnel deployment time cost among all plans is 60 minutes and the maximum is 180 minutes, and the time cost of a certain plan is 90 minutes, then its normalized value is (90-60)÷(180-60)=0.25. Through normalization, these three indicators can be calculated and compared on the same scale.

[0118] Step S453: Based on the objective function values ​​of the normalized total time cost of personnel deployment, the load balance of candidate maintenance sub-centers, and the skill matching degree of maintenance personnel, calculate the fitness values ​​of several maintenance personnel deployment schemes as the comprehensive evaluation value of each scheme.

[0119] The fitness value of each maintenance personnel allocation plan is calculated based on the objective function values ​​of the normalized total time cost of personnel allocation, the load balance of candidate maintenance sub-centers, and the skill matching degree of maintenance personnel. This fitness value is used as the comprehensive evaluation value of the plan.

[0120] Step S454: Select the maintenance personnel allocation plan with the highest comprehensive evaluation value and allocate maintenance personnel accordingly.

[0121] After calculating the comprehensive evaluation value of all maintenance personnel allocation plans, the system automatically selects the plan with the highest comprehensive evaluation value. This plan achieves an optimal balance in three key aspects: total time cost of personnel allocation, load balancing of candidate maintenance centers, and skill matching of maintenance personnel. For example, among multiple plans, one plan has a comprehensive evaluation value of 0.7, higher than the others, which means that this plan performs best in terms of time cost control, load balancing of centers, and personnel skills meeting task requirements. Selecting this plan for actual maintenance personnel allocation can ensure the efficient and high-quality completion of maintenance tasks to the greatest extent, while also taking into account the long-term operational stability of the centers and achieving rational resource allocation.

[0122] Furthermore, in step S453, based on the objective function values ​​of the normalized total time cost of personnel deployment, the load balance of candidate maintenance sub-centers, and the skill matching degree of maintenance personnel, fitness values ​​for several maintenance personnel deployment schemes are calculated, including:

[0123] Step S453a: Based on the objective function values ​​of the normalized total time cost of personnel allocation, the load balance of candidate maintenance sub-centers, and the skill matching degree of maintenance personnel, construct the original vector of each maintenance personnel allocation plan.

[0124] Based on the normalized objective function values ​​(total time cost of personnel deployment, load balancing of candidate maintenance sub-centers, and skill matching degree of maintenance personnel), an original vector is constructed for each maintenance personnel deployment plan. Normalization unifies the objective function values ​​to the range [0,1], ensuring comparability of each objective in the comprehensive evaluation. Each dimension of the original vector corresponds to a normalized objective function value. For example, for a maintenance personnel deployment plan, its original vector can be represented as [T′,E′,F′], where T′ is the normalized total time cost of personnel deployment, E′ is the normalized load balancing degree, and F′ is the normalized skill matching degree. The original vector serves as the basic input data for the multi-objective optimization problem, used for subsequent feature extraction and fitness value calculation.

[0125] Step S453b: Input the original vector into the pre-trained autoencoder model to obtain the low-dimensional feature encoding vector.

[0126] The original vector is input into a pre-trained autoencoder model to obtain a low-dimensional feature encoding vector. An autoencoder is an unsupervised learning model commonly used for dimensionality reduction and feature extraction. It learns the latent representation of data by compressing the input data to a low-dimensional space (encoding) and then reconstructing it back into the original space (decoding). The pre-trained autoencoder model has been trained on historical data and can effectively extract low-dimensional features from the original vector. The low-dimensional feature encoding vector better reflects the latent structure of the original data while reducing data dimensionality and improving computational efficiency. For example, the original vector [T′,E′,F′] may be compressed into a 2-dimensional feature encoding vector [f1,f2] after passing through an autoencoder.

[0127] Step S453c: Input the low-dimensional feature encoding vector into the pre-trained support vector regression model, and obtain the fitness value of the maintenance personnel transfer plan based on the support vector regression model.

[0128] The low-dimensional feature encoding vector is input into a pre-trained Support Vector Regression (SVR) model, and the fitness value of the maintenance personnel deployment plan is obtained based on the SVR model. Support Vector Regression predicts continuous output values ​​by fitting the relationship between input features and target values. The pre-trained SVR model has been trained using historical data (including low-dimensional feature encoding vectors and corresponding fitness values) and can predict fitness values ​​based on low-dimensional features. For example, after the low-dimensional feature encoding vector [f1, f2] is input into the SVR model, the output fitness value reflects the overall performance of the maintenance personnel deployment plan and is used to compare and select the optimal plan.

[0129] Specifically, the formula for calculating the fitness value of the maintenance personnel allocation plan in step S450 is as follows:

[0130]

[0131] Where T′ is the normalized value of the total time cost of personnel deployment, E′ is the normalized value of the load balance of the candidate maintenance sub-center, F′ is the normalized value of the skill matching degree of maintenance personnel, and α, β, and λ are the corresponding coefficients.

[0132] The fitness value calculation formula above evaluates the merits of different maintenance personnel allocation schemes by comprehensively considering three key factors: the normalized total time cost of personnel allocation, the load balance of candidate maintenance sub-centers, and the skill matching degree of maintenance personnel.

[0133] First, the formula introduces weights α, β, and λ, corresponding to time cost, load balancing, and skill matching, respectively. This weighted approach reflects the relative importance of different factors in the decision-making process. For example, if time cost is more critical in practical applications, then α can be given a higher weight. The weight setting allows the fitness function to flexibly adapt to different practical needs and priorities, thus more accurately reflecting the overall merits of the solution.

[0134] Secondly, the formula uses a reciprocal form, which normalizes the fitness value to the range [0,1], facilitating comparison and ranking between different solutions. The closer the fitness value is to 1, the better the solution; while the closer it is to 0, the worse the overall performance of the solution. This normalization method makes the fitness value interpretable and comparable, making it easy to quickly identify the optimal solution.

[0135] In handling specific factors, lower time costs are better, so T′ is directly used as a positive influencing factor. Load balance and skill matching, on the other hand, are better the higher they are. Therefore, (1-E′) and (1-F′) are used in the formula to represent the negative impact on fitness, ensuring logical consistency: when a factor performs poorly, the negative impact on fitness is reflected by increasing the value in the denominator, thus lowering the fitness value; conversely, when a factor performs well, the negative impact on fitness is smaller, and the fitness value is relatively higher.

[0136] In addition, the "1" in the formula acts as a base value in the denominator, avoiding the case where the denominator is zero. It also provides a baseline for the fitness value. Even if all factors perform very poorly, the fitness value will not be zero, but will approach zero. This is reasonable in practical applications because even if the overall performance of the scheme is poor, it is still feasible to some extent, but it needs further optimization.

[0137] The above calculation formula, by reasonably integrating multiple key factors and adopting a weighted and reciprocal approach, can not only comprehensively evaluate the advantages and disadvantages of different maintenance personnel allocation plans, but also flexibly adapt to different actual needs and priorities.

[0138] In an optional embodiment of the present invention, the functional formula for the total time cost of personnel deployment is as follows:

[0139]

[0140] Where n is the number of candidate maintenance sub-centers, m is the number of maintenance personnel levels, and n ij d represents the number of level j maintenance personnel to be drawn from the i-th candidate maintenance sub-center. The more personnel drawn, the higher the corresponding time cost. iLet v be the distance between the i-th candidate maintenance sub-center and the location where the maintenance task is to be carried out. The farther the distance, the longer the time required for personnel to arrive, and the higher the time cost. j Let be the average movement speed of maintenance personnel at level j. Maintenance personnel at different levels may have different movement speeds due to factors such as transportation equipment, skills, and experience. The faster the speed, the shorter the time to reach the task location, and the lower the time cost. The above formula can be used to calculate the total time cost required to dispatch personnel from multiple candidate maintenance sub-centers to complete a maintenance task order.

[0141] The total time cost of personnel deployment is normalized to obtain the normalized value of the total time cost of personnel deployment:

[0142]

[0143] Among them, T max T represents the maximum total time cost of personnel deployment across all maintenance personnel allocation plans. min Minimize the total time cost of personnel allocation in all maintenance personnel allocation plans.

[0144] Through this calculation, the normalized value T′ of the total time cost of personnel deployment is obtained, ranging from 0 to 1. If T′ of a certain plan is close to 0, it means that the total time cost of personnel deployment for that plan is relatively low compared to all plans; conversely, if T′ is close to 1, it means that the time cost of that plan is relatively high.

[0145] In an optional embodiment of the present invention, the functional formula for the load balance of candidate maintenance sub-centers is as follows:

[0146]

[0147]

[0148] Where n is the number of candidate maintenance sub-centers, m is the number of maintenance personnel levels, and n ij L represents the number of level j maintenance personnel drawn from the i-th candidate maintenance sub-center. i For the original workload of the i-th candidate maintenance sub-center, ω j The workload weight assigned to the j-th level maintenance personnel R represents the load increase value after adding the current maintenance task to the i-th candidate maintenance sub-center. i Add the relative load value after the current maintenance tasks to each candidate maintenance sub-center. This represents the average relative load of each candidate maintenance sub-center.

[0149] In the above formula, The increased workload of the i-th candidate maintenance sub-center due to personnel reassignment is calculated to quantify the impact of personnel reassignment on the existing workload of the maintenance sub-center, while also reflecting the resource consumption of different skill levels. R i Normalization eliminates differences in the size of sub-centers, facilitating horizontal comparison of load distribution. The standard deviation is used to measure the dispersion of the relative load of each maintenance sub-center, which is used to quantify the fairness of resource allocation when multiple sub-centers collaborate, and to avoid local overload.

[0150] The load balance of the candidate maintenance sub-centers is normalized to obtain the normalized value of the load balance of the candidate maintenance sub-centers:

[0151]

[0152] Among them, E max E represents the maximum load balance of candidate maintenance sub-centers in all maintenance personnel allocation plans. min The minimum load balance of candidate maintenance sub-centers in the plan for all maintenance personnel allocation.

[0153] By normalizing, the load balance degree E is mapped to the interval [-1,1], which facilitates comprehensive evaluation with other indicators (time cost, skill matching degree); the influence of the absolute value of the original standard deviation is eliminated, and only the relative superiority and inferiority relationship is retained. Since a smaller E indicates a higher degree of balance, and a larger E′ value after normalization indicates a worse degree of balance, it is necessary to combine the weights in the comprehensive evaluation formula (such as using 1-E′) for positive transformation.

[0154] In an optional embodiment of the present invention, the functional formula for the skill matching degree of maintenance personnel is as follows:

[0155]

[0156] Where n is the number of candidate maintenance sub-centers, m is the number of maintenance personnel levels, and n ij Let F be the number of level j maintenance personnel to be drawn from the i-th candidate maintenance sub-center, k be the k-th skill required for the current maintenance task, W be the number of required skills, and F be the number of skills required. i S represents the skill matching degree of the personnel to be drawn from the i-th candidate maintenance sub-center. k M represents the standard requirement value for the k-th skill that needs to be mastered. ik Let q be the average mastery level of the j-th level maintenance personnel selected from the i-th candidate maintenance sub-center for the k-th skill. k This represents the weight of skill k for level j maintenance personnel in the current maintenance task.

[0157] The skill matching degree of maintenance personnel is normalized to obtain the normalized value of skill matching degree of maintenance personnel:

[0158]

[0159] Among them, F max F represents the maximum skill matching degree of maintenance personnel in all maintenance personnel allocation plans. min The minimum skill matching value for maintenance personnel in the allocation plan.

[0160] The normalized F′ facilitates comparison and comprehensive analysis with other evaluation indicators (such as total time cost of personnel deployment and load balance of candidate maintenance sub-centers) on the same scale. By mapping skill matching to a unified interval, the relative advantages and disadvantages of different deployment schemes in terms of skill matching can be more clearly determined, providing more effective data support for the final selection of the optimal maintenance personnel deployment scheme. For example, during comprehensive evaluation, appropriate weights can be assigned to F′ according to different business needs, and the comprehensive evaluation value can be calculated together with other normalized indicators, thereby making a more reasonable decision.

[0161] Accordingly, please refer to Figure 2 A second aspect of this invention provides a maintenance support and assurance management system, which manages maintenance task orders based on the above-described maintenance support and assurance management method, including:

[0162] Task receiving module 1 is used to receive maintenance task orders and obtain maintenance task location information and maintenance task level information from the maintenance task orders;

[0163] Node selection module 2 is used to obtain several maintenance sub-centers within a preset range of the maintenance task location that meet the maintenance task level requirements and whose current workload load value is lower than the preset workload threshold as candidate maintenance sub-centers.

[0164] Personnel allocation module 3 is used to select one of the candidate maintenance sub-centers to allocate maintenance personnel to complete the maintenance task order when the available maintenance personnel of at least one candidate maintenance sub-center meet the personnel requirements of the maintenance task level.

[0165] The personnel allocation module 3 is also used to allocate maintenance personnel from multiple candidate maintenance sub-centers to jointly complete the maintenance task order when the number of available maintenance personnel in each candidate maintenance sub-center is less than the personnel requirements for the maintenance task level. The allocated maintenance personnel from multiple candidate maintenance sub-centers meet the personnel requirements for the maintenance task level.

[0166] Accordingly, a third aspect of the present invention provides an electronic device, including: at least one processor; and a memory connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to cause the at least one processor to perform the above-described maintenance support and assurance management method.

[0167] Accordingly, a fourth aspect of the present invention provides a computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the above-described maintenance support and assurance management method.

[0168] This invention aims to protect a maintenance support and assurance management method and system. The management method includes the following steps: receiving maintenance task orders; obtaining maintenance task location information and maintenance task level information from the maintenance task orders; obtaining several maintenance sub-centers within a preset range of the maintenance task location that meet the maintenance task level requirements and whose current workload is below a preset workload threshold as candidate maintenance sub-centers; when the available maintenance personnel at at least one candidate maintenance sub-center meet the personnel requirements for the maintenance task level, selecting one candidate maintenance sub-center to allocate maintenance personnel to complete the maintenance task order; when the available maintenance personnel at each candidate maintenance sub-center are less than the personnel requirements for the maintenance task level, allocating maintenance personnel from multiple candidate maintenance sub-centers to jointly complete the maintenance task order, wherein the allocated maintenance personnel from the multiple candidate maintenance sub-centers meet the personnel requirements for the maintenance task level. The above technical solution has the following effects:

[0169] 1. Through a dynamic screening mechanism and a hierarchical matching strategy, precise positioning and rapid response of maintenance resources are achieved. The preset range of task locations is defined, and candidate sub-centers are screened based on their qualifications, skills, and load thresholds to ensure basic matching. When allocating resources to a single sub-center, a dual strategy of "distance priority" or "load priority" is adopted: emergency tasks prioritize calling nearby sub-centers to shorten response time, while routine tasks avoid local overload through a load balancing formula. The dynamic switching mechanism effectively solves the contradiction between resource idleness and overload in traditional scheduling, increasing sub-center utilization by 20%-30% and reducing the average task response time by 15%-25%.

[0170] 2. A multi-dimensional comprehensive evaluation model was used to achieve globally optimal decision-making in complex scenarios. When multiple sub-centers jointly coordinate, the system generates multiple allocation plans and calculates the normalized values ​​of time cost, load balance, and skill matching degree, and performs a comprehensive score. Compared with traditional single-index scheduling, this model improves the overall optimization rate of the plan by 40%-50% and the completion efficiency of cross-sub-center collaborative tasks by more than 35%, ensuring the coordinated optimization of task response speed and execution quality.

[0171] 3. Through quantitative models and normalization, the system achieves standardization and scalability in scheduling decisions. The time cost formula incorporates personnel movement speed and distance into the calculation, ensuring time comparability; the load balance formula quantifies the fairness of resource allocation through standard deviation; the skill matching formula combines skill weights and mastery levels to ensure precise capability alignment; and the normalization process eliminates differences in indicator dimensions, supporting dynamic adjustment of weight parameters. This quantitative system can quickly adapt to the actual needs of equipment maintenance and repair management, providing a standardized framework for multi-objective decision-making and significantly enhancing the intelligence level and overall efficiency of equipment maintenance services.

[0172] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0173] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0174] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0175] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1The steps of the function specified in one or more boxes.

[0176] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A maintenance support and assurance management method, characterized in that, Includes the following steps: Receive maintenance task orders and obtain the maintenance task location information and maintenance task level information of the maintenance task orders; Several maintenance sub-centers within a preset range of the maintenance task location that meet the maintenance task level requirements and whose current workload is lower than a preset workload threshold are selected as candidate maintenance sub-centers. When at least one of the candidate maintenance sub-centers has available maintenance personnel to meet the personnel requirements of the maintenance task level, one of the candidate maintenance sub-centers' maintenance personnel shall be selected and assigned to complete the maintenance task order; When the number of available maintenance personnel at each of the candidate maintenance sub-centers is less than the personnel requirement for the maintenance task level, maintenance personnel from multiple candidate maintenance sub-centers are deployed to jointly complete the maintenance task order, and the deployed maintenance personnel from multiple candidate maintenance sub-centers meet the personnel requirement for the maintenance task level. The process of coordinating maintenance personnel from multiple candidate maintenance sub-centers to jointly complete the maintenance task order includes: Obtain the distance values ​​between the candidate maintenance sub-centers and the location where the maintenance task is carried out; Obtain the number of available maintenance personnel at each level in the multiple candidate maintenance sub-centers; Based on the distance value and the number of available maintenance personnel at each level, select several of the multiple candidate maintenance sub-centers to obtain several maintenance personnel allocation plans. Obtain the total time cost of personnel allocation, the load balance of candidate maintenance sub-centers, and the skill matching degree of maintenance personnel for each of the aforementioned maintenance personnel allocation plans; Based on the total time cost of personnel deployment, the load balance of the candidate maintenance sub-centers, and the skill matching degree of the maintenance personnel, a comprehensive evaluation value is calculated for each of the maintenance personnel deployment schemes, and maintenance personnel are deployed according to the maintenance personnel deployment scheme corresponding to the optimal result of the comprehensive evaluation value. Based on the distance value and the number of available maintenance personnel at each level, several candidate maintenance sub-centers are selected to obtain several maintenance personnel allocation plans, including: Randomly select several of the candidate maintenance sub-centers; Obtain the skill matching degree between each candidate maintenance sub-center and the maintenance task order, and construct a distance matrix for the currently randomly selected candidate maintenance sub-center by combining the distance value and the number of available maintenance personnel at each level; Based on the hierarchical clustering algorithm, the candidate maintenance sub-centers are clustered, and according to the maintenance task order, a number of randomly selected candidate maintenance sub-centers are divided into a main node cluster and a number of sub-node clusters. The candidate maintenance sub-center that is closest to the location where the maintenance task is implemented in the main node cluster is taken as the main node, and the candidate maintenance sub-center that is closest to the main node in each sub-node cluster is taken as the sub-node. Based on the main node and the candidate maintenance sub-centers corresponding to the sub-nodes, a maintenance personnel allocation plan is determined, and several maintenance personnel allocation plans are obtained after multiple random selections. The comprehensive evaluation value for each maintenance personnel allocation plan is calculated based on the total time cost of personnel deployment, the load balance of the candidate maintenance sub-centers, and the skill matching degree of the maintenance personnel, including: Based on a multi-objective optimization model, an objective function is constructed for each of the maintenance personnel allocation schemes. The objective function includes: minimizing the total time cost of personnel allocation, maximizing the load balance of candidate maintenance sub-centers, and maximizing the skill matching degree of maintenance personnel. Based on the objective function of the maintenance personnel allocation plan, the objective function values ​​of the total time cost of personnel allocation, the load balance of the candidate maintenance sub-centers, and the skill matching degree of the maintenance personnel are calculated and normalized respectively. Based on the objective function values ​​of the total time cost of personnel deployment after normalization, the load balance of the candidate maintenance sub-centers, and the skill matching degree of the maintenance personnel, the fitness values ​​of several maintenance personnel deployment schemes are calculated as the comprehensive evaluation value of each scheme. The maintenance personnel allocation plan with the highest comprehensive evaluation value was selected for allocation.

2. The maintenance support and assurance management method according to claim 1, characterized in that, The maintenance task levels include: Level 1 maintenance task, Level 2 maintenance task, and Level 3 maintenance task; the maintenance personnel levels include: junior, intermediate, senior, and expert. The personnel requirements for the Level 1 maintenance task, the Level 2 maintenance task, and the Level 3 maintenance task correspond to the corresponding preset number of junior, intermediate, senior, and / or expert level maintenance personnel, respectively.

3. The maintenance support and assurance management method according to claim 1 or 2, characterized in that, The step of selecting and assigning maintenance personnel from the candidate maintenance sub-center to complete the maintenance task order includes: Obtain the current workload load value of the candidate maintenance sub-centers, and select the candidate maintenance sub-centers in ascending order of the current workload load value.

4. The maintenance support and assurance management method according to claim 3, characterized in that, The current workload value The calculation formula is: ； in, For maintenance task level values, The work efficiency of the j-th maintenance personnel level in the current candidate maintenance sub-center. Let j be the number of available slots for the j-th maintenance personnel level. Let be the distance between the current candidate maintenance sub-center and the h-th maintenance task location. The distance attenuation coefficient between the maintenance task location and the candidate maintenance sub-center. These are the coefficients of the cosine function.

5. The maintenance support and assurance management method according to claim 1, characterized in that, The fitness values ​​of several maintenance personnel allocation schemes are calculated based on the objective function values ​​of the normalized total time cost of personnel deployment, the load balance of the candidate maintenance sub-centers, and the skill matching degree of the maintenance personnel, including: Based on the objective function values ​​of the normalized total time cost of personnel deployment, the load balance of the candidate maintenance sub-centers, and the skill matching degree of the maintenance personnel, the original vector of each maintenance personnel deployment scheme is constructed. The original vector is input into a pre-trained autoencoder model to obtain a low-dimensional feature encoding vector; The low-dimensional feature encoding vector is input into a pre-trained support vector regression model, and the fitness value of the maintenance personnel deployment scheme is obtained based on the support vector regression model.

6. The maintenance support and assurance management method according to claim 1, characterized in that, The formula for calculating the fitness value of the maintenance personnel allocation plan is as follows: ; in, The normalized value of the total time cost of personnel deployment. The normalized value for load balance of candidate maintenance sub-centers. To normalize the skill matching degree of maintenance personnel, , , For the corresponding coefficient.

7. A maintenance support and assurance management system, characterized in that, The maintenance support and assurance management method according to any one of claims 1-6 manages maintenance task orders, including: The task receiving module is used to receive maintenance task orders and obtain the maintenance task location information and maintenance task level information of the maintenance task orders. The node selection module is used to obtain several maintenance sub-centers within a preset range of the maintenance task location that meet the maintenance task level requirements and whose current workload load value is lower than the preset workload threshold as candidate maintenance sub-centers. The personnel allocation module is used to select one of the candidate maintenance sub-centers to allocate maintenance personnel to complete the maintenance task order when at least one of the candidate maintenance sub-centers has available maintenance personnel to meet the personnel requirements of the maintenance task level. The personnel allocation module is also used to allocate maintenance personnel from multiple candidate maintenance sub-centers to jointly complete the maintenance task order when the number of available maintenance personnel in each candidate maintenance sub-center is less than the personnel requirement for the maintenance task level. The allocated maintenance personnel from multiple candidate maintenance sub-centers meet the personnel requirement for the maintenance task level.

Images