An operation and maintenance personnel matching processing method and system, and a storage medium

Abstract

The application relates to the technical field of information technology service management, and discloses an operation and maintenance personnel ratio processing method and system and a storage medium, the method comprising the following steps: obtaining work order performance data, business load data, value-keeping demand data and automation coverage data of a target operation and maintenance object in a preset statistical period; determining a basic configuration number according to the business load data; generating a risk adjustment factor according to the work order performance data; determining a value-keeping additional number according to the value-keeping demand data, determining an automation reduction number according to the automation coverage data, and generating a preliminary establishment number; performing constraint calibration on the preliminary establishment number to obtain a target establishment number; and dynamically adjusting the target establishment number according to the corresponding relationship between the risk adjustment factor and a preset risk threshold interval and in combination with a continuous statistical period determination result. The application improves the accuracy, stability and executability of the operation and maintenance personnel ratio result.

Description

Technical Field

[0001] This invention relates to the field of information technology service management technology, and more specifically, to a method, system, and storage medium for processing the allocation of maintenance personnel. Background Technology

[0002] With the continuous expansion of government and enterprise informatization and industrial information systems, the scope of operation and maintenance (O&M) has evolved from single equipment maintenance to a comprehensive service system encompassing business systems, infrastructure, work order processes, and on-call support. O&M organizations not only need to ensure fault response and business continuity but also need to balance personnel input, service quality, and resource utilization efficiency. Against this backdrop, how to rationally allocate O&M personnel based on operational data has become a key issue in IT service management and resource planning.

[0003] In existing technologies, such as the invention patent CN107358330A – a method for configuring special job personnel based on a predictor – the solution mainly establishes a personnel demand model and configures personnel based on data such as equipment composition and quantity, line length, maintenance hours, and existing personnel configuration. This can solve the personnel configuration calculation problem in certain scenarios. However, its examples also reflect that existing personnel configuration solutions rely more on scale, hours, or historical configuration for total modeling. They often lack more detailed linkage processing for changes in personnel demand caused by fluctuations in performance pressure, differences in duty requirements, skill coverage requirements, and changes in automation capabilities in maintenance scenarios. This results in configuration results that are prone to being static and difficult to continuously adapt to actual changes in the maintenance service process.

[0004] Therefore, it is necessary to design a method, system, and storage medium for managing the allocation of maintenance personnel to solve the problems existing in the current technology. Summary of the Invention

[0005] In view of this, the present invention proposes a method, system and storage medium for processing the allocation of operation and maintenance personnel, aiming to solve the problem that the existing operation and maintenance personnel allocation scheme mainly relies on static scale or historical configuration for calculation, resulting in insufficient matching between the allocation results and actual operation and maintenance needs.

[0006] This invention proposes a method for handling the allocation ratio of operation and maintenance personnel, including: Acquire work order fulfillment data, business load data, duty requirement data, and automation coverage data of the target maintenance object within a preset statistical period; The basic configuration number is determined based on the business load data, and a risk adjustment factor is generated based on the work order fulfillment data; the additional number of on-duty personnel is determined based on the on-duty requirement data, and the number of automation staff reductions is determined based on the automation coverage data; and a preliminary staffing number is generated based on the basic configuration number, the risk adjustment factor, the additional on-duty personnel number, and the number of automation staff reductions. The target staffing number is obtained by constraining and calibrating the initial staffing number. The constraints include duty coverage constraints, skill coverage constraints, performance constraints, and staffing fluctuation constraints. Based on the correspondence between the risk adjustment factor and the preset risk threshold range, and in conjunction with the judgment results of continuous statistical periods, the target number of compilations is dynamically adjusted.

[0007] Furthermore, when generating risk adjustment factors based on the work order performance data, the following are included: Within a sliding time window, the risk components of event work orders and problem work orders are determined respectively. The risk components of event work orders and problem work orders are then aggregated based on the number of event work orders, the number of problem work orders, and the priority distribution to generate the risk adjustment factor.

[0008] Furthermore, when determining the risk components of event work orders and problem work orders respectively, the following are included: The deviation values ​​of event work orders and problem work orders are determined based on the timeliness of response, timeliness of closure, reopening, and backlog, respectively. The deviation values ​​of event work orders and problem work orders are then weighted to obtain the risk components of event work orders and problem work orders.

[0009] Furthermore, when constraining and calibrating the preliminary staffing number to obtain the target staffing number, the following steps are included: Using the initial staffing number as the initial value, the duty coverage constraint, the skill coverage constraint, the performance constraint, and the staffing fluctuation constraint are checked in sequence. The initial staffing number that does not meet the constraints is corrected until the target staffing number that meets all constraints is obtained.

[0010] Furthermore, the continuous statistical period determination results include: if the risk adjustment factor is continuously higher than the high-risk threshold for at least two consecutive statistical periods, it is determined to be continuously high-risk; if the risk adjustment factor is continuously lower than the low-risk threshold, it is determined to be continuously low-risk; if the risk adjustment factor is in the hysteresis interval, it is determined to be in a state of sustained volatility.

[0011] Furthermore, based on the state of sustained high risk, sustained low risk, or continued fluctuation, the target number of staffing is increased, decreased, or kept unchanged, and the dynamically adjusted target number of staffing is used as the reference number for the next preset statistical period.

[0012] Furthermore, when increasing the target staffing level based on the sustained high risk, the target staffing level is increased according to a preset adjustment step size; when decreasing the target staffing level based on the sustained low risk, the target staffing level is decreased according to the preset adjustment step size.

[0013] Furthermore, when the risk adjustment factor changes from a state of sustained high risk or sustained low risk to a state of sustained volatility, the target compilation number of the previous preset statistical period is continued.

[0014] Compared with existing technologies, the beneficial effects of this invention are as follows: By simultaneously acquiring data within a preset statistical period and forming basic configuration numbers, risk adjustment factors, on-call additional numbers, and automation attrition numbers, and then generating preliminary staffing numbers based on these, and performing constraint calibration and dynamic adjustment, the personnel allocation results no longer rely solely on static calculations based on a single scale, working hours, or historical configuration. Instead, they comprehensively reflect changes in service pressure, on-call responsibilities, skill coverage requirements, and automation substitution capabilities in actual operation and maintenance scenarios, thereby improving the matching degree between the staffing results and actual operation and maintenance needs. Furthermore, generating risk adjustment factors based on work order fulfillment data facilitates the incorporation of fault response and service fulfillment status into the personnel allocation process. This approach ensures that the staffing results can promptly reflect changes in operational pressure. By calibrating the initial staffing numbers through constraints on duty coverage, skill coverage, contract fulfillment, and staffing fluctuations, it avoids neglecting issues such as job coverage, contract fulfillment capabilities, and adjustment stability when calculating solely from a quantitative perspective. This makes the obtained target staffing numbers more achievable and applicable to engineering projects. Furthermore, by dynamically adjusting the target staffing numbers based on the correspondence between risk adjustment factors and preset risk threshold ranges, as well as the results of continuous statistical cycles, it can suppress frequent increases and decreases in staffing caused by short-term fluctuations, improving the stability and continuous adaptability of staffing adjustments. This solution ensures the quality of operation and maintenance services while also considering the rationality of personnel input and the efficiency of resource utilization.

[0015] On the other hand, this application also provides a system for processing the allocation of maintenance personnel, used to apply the above-mentioned method for processing the allocation of maintenance personnel, including: The data acquisition unit is used to acquire work order fulfillment data, business load data, duty requirement data, and automation coverage data of the target maintenance object within a preset statistical period. The processing unit is used to determine the basic configuration number based on the business load data, generate a risk adjustment factor based on the work order fulfillment data, determine the additional duty personnel number based on the duty requirement data, determine the automation reduction number based on the automation coverage data, and generate a preliminary staffing number based on the basic configuration number, the risk adjustment factor, the additional duty personnel number, and the automation reduction number. The calibration unit is used to perform constraint calibration on the preliminary staffing number to obtain the target staffing number. The constraints include duty coverage constraints, skill coverage constraints, performance constraints, and staffing fluctuation constraints. The adjustment unit is used to dynamically adjust the target number of compilations based on the correspondence between the risk adjustment factor and the preset risk threshold range, and in combination with the judgment results of the continuous statistical period.

[0016] On the other hand, this application also provides a computer-readable storage medium in which computer-executable instructions stored in the computer-readable storage medium are executed by a processor of an electronic device, enabling the electronic device to perform the above-described method.

[0017] It is understandable that the above-mentioned methods for allocating maintenance personnel, systems, and storage media have the same beneficial effects, and will not be elaborated further here. Attached Figure Description

[0018] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart of the operation and maintenance personnel allocation method provided in the embodiments of the present invention; Figure 2 This is a functional block diagram of the maintenance personnel allocation system provided in an embodiment of the present invention. Detailed Implementation

[0019] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] In some embodiments of this application, see Figure 1 As shown, this application proposes a method for handling the allocation of operation and maintenance personnel, including: S100: Obtain work order fulfillment data, business load data, duty requirement data, and automation coverage data of the target maintenance object within a preset statistical period.

[0021] S200: Determine the basic configuration number based on workload data, and generate a risk adjustment factor based on work order fulfillment data. Determine the additional on-call staff number based on on-call demand data, and determine the automation staff reduction number based on automation coverage data. Generate a preliminary staffing number based on the basic configuration number, risk adjustment factor, additional on-call staff number, and automation staff reduction number.

[0022] S300: Constraint calibration is performed on the initial staffing number to obtain the target staffing number. The constraints include duty coverage constraints, skill coverage constraints, performance constraints, and staffing fluctuation constraints.

[0023] S400: Based on the correspondence between the risk adjustment factor and the preset risk threshold range, and combined with the judgment results of the continuous statistical period, the target number is dynamically adjusted.

[0024] Specifically, in step S100, the preset statistical period is preferably set to one month to form a complete compilation calculation and adjustment result. Work order fulfillment data is used to characterize service fulfillment pressure, including at least the timeliness of response, closure timeliness, reopening status, backlog status, and priority distribution of event work orders and problem work orders. Business load data is used to characterize the service scale and workload of the current maintenance objects, including at least the number of active users, total number of devices, number of coverage nodes, and cross-regional support scope. On-call duty requirement data is used to characterize shift guarantee requirements, including at least the on-call duty mode, number of shifts, and time period coverage requirements. Automation coverage data is used to characterize the degree to which automation tools replace manual work, including at least the automatic inspection coverage rate, automatic repair coverage rate, and batch operation coverage rate. After collection, abnormal missing values ​​are corrected, duplicate records across systems are deduplicated, and data from different sources are uniformly mapped to the same statistical period.

[0025] In step S200, the basic configuration number is determined based on the business load data. The basic configuration number represents the minimum configuration level required to maintain basic operation and maintenance services without considering short-term fluctuations. In this embodiment, it can be obtained by taking the larger value of the user-side baseline and the device-side baseline. Specifically, for the first-level headquarters, the user-side baseline can be calculated by dividing the number of active users by 500, and the device-side baseline can be calculated by dividing the number of managed devices by 1000. For second-level units, the baselines are calculated by dividing the number of active users by 100 and the number of devices by 200, respectively. Subsequently, a risk adjustment factor is generated based on the work order fulfillment data. To make indicators of different dimensions comparable, each work order fulfillment indicator is first normalized, and the deviation value is calculated. For indicators such as response timeliness and closure timeliness, where a larger value indicates a better fulfillment status, the deviation value can be calculated by subtracting the actual value from the target value and then dividing by the target value. For indicators such as reopening rate, backlog rate, and average resolution time, where higher values ​​indicate worse performance, the deviation can be calculated by subtracting the target value from the actual value and then dividing by the target value; a deviation less than 0 is taken as 0. Based on this, the risk components of event work orders and problem work orders can be written as follows: Event work order risk component = event response deviation value × event response weight + event closure deviation value × event closure weight + event reopening deviation value × event reopening weight + event backlog deviation value × event backlog weight.

[0026] Problem work order risk component = Problem response deviation value × Problem response weight + Problem closure deviation value × Problem closure weight + Problem reopening deviation value × Problem reopening weight + Problem backlog deviation value × Problem backlog weight.

[0027] After obtaining the two risk components, a weighted aggregation is performed based on the number of event work orders, the number of problem work orders, and their priority distribution to obtain a risk adjustment factor. Higher priority work orders receive greater weights to ensure their higher contribution to the risk adjustment factor. Subsequently, the additional staffing numbers are determined based on staffing requirements data, and the automation reduction numbers are determined based on automation coverage data. A preliminary staffing plan is then generated according to the following relationship: Preliminary staffing count = Basic configuration count + Risk adjustment count + On-duty additional count - Automation staff reduction count.

[0028] The risk correction number is obtained by mapping the base configuration number to the risk adjustment factor; the higher the risk adjustment factor, the larger the risk correction number. The risk correction number can be determined segmented according to the range of the risk adjustment factor. When the risk adjustment factor is below the low-risk threshold, the risk correction number is 0. When the risk adjustment factor is between the low-risk and high-risk thresholds, the risk correction number is determined by multiplying the base configuration number by a first correction ratio. When the risk adjustment factor is above the high-risk threshold, the risk correction number is determined by multiplying the base configuration number by a second correction ratio, where the second correction ratio is greater than the first correction ratio. Preferably, the first correction ratio can be 5% to 15%, and the second correction ratio can be 15% to 35%, and can be calibrated based on the risk adjustment factor distribution and performance / default results of the same-level target maintenance objects over the past 12 statistical periods. The on-duty supplement number is used to compensate for the manpower required for continuous on-duty work, nighttime support, and cross-regional collaboration. The automation reduction number is used to offset the substitution of manual work by automatic inspection, fault self-healing, and batch processing. The on-duty supplement number can be determined according to the on-duty mode and segmented rules. Preferably, in a 24 / 7 on-call scenario, the additional on-call personnel for the primary headquarters can be 40% to 60% of the basic configuration, and for secondary units, it can be 30% to 50%. In a fixed-period on-call scenario during weekdays, the additional on-call personnel is 0. In an emergency on-call scenario during nighttime and holidays, the additional on-call personnel can be 10% to 25% of the basic configuration. The number of automated personnel reductions can be determined in a tiered manner based on the effective automation coverage rate. Preferably, for every 20 percentage point increase in effective automation coverage rate, the number of automated personnel reductions increases by 0.1 to 0.3 people, with the maximum automated personnel reduction for the primary headquarters not exceeding 1.0 person and the maximum automated personnel reduction for secondary units not exceeding 0.8 people.

[0029] In step S300, the preliminary staffing number is not directly used as the final result. Instead, constraint calibration is performed on the preliminary staffing number to obtain the target staffing number. The duty coverage constraint ensures that each shift is staffed under the established duty mode. The skill coverage constraint ensures that key skill domains meet at least the minimum staffing requirements. The performance constraint ensures that, under current work order pressure, the on-time response rate, on-time closure rate, and average resolution time meet the preset service level agreement requirements. The staffing fluctuation constraint limits the magnitude of changes in a single adjustment relative to the previous statistical period. In practice, the preliminary staffing number can be used as a candidate value, and each constraint is checked step by step. Candidate values ​​that do not meet the constraints are corrected according to a preset step size until all constraints are met.

[0030] In step S400, a preset risk threshold range and continuous statistical period judgment results are set. The high-risk threshold, low-risk threshold, and hysteresis interval are preferably obtained through historical sample calibration. Specifically, they can be determined by the mapping relationship between the risk adjustment factor and the actual default rate of the same level unit over the past 12 statistical periods. For example, the dividing point where the probability of default increases significantly can be set as the high-risk threshold, and the upper limit of the long-term stable compliance range can be set as the low-risk threshold; the interval between the two is defined as the hysteresis interval. The continuous statistical period judgment results are used to filter occasional spikes and fluctuations, and are preferably set to trigger an action only after two consecutive statistical periods of consistent status. When the risk adjustment factor is higher than the high-risk threshold for two consecutive statistical periods, it is judged as a continuous high-risk situation and an increase in allocation is triggered. When the risk adjustment factor is lower than the low-risk threshold for two consecutive statistical periods, it is judged as a continuous low-risk situation and a reduction in allocation is triggered. When the risk adjustment factor is within the hysteresis interval, or when it transitions from a continuous high-risk state or a continuous low-risk state into the hysteresis interval, the current target allocation remains unchanged. The reason for this setup is that sudden failures, concentrated releases, or temporary events within a single statistical period may cause short-term anomalies in performance indicators. Directly increasing or decreasing allocations could easily lead to fluctuations in the performance evaluation. By using continuous statistical periods for assessment, adjustment stability can be improved without sacrificing responsiveness.

[0031] Specifically, in this embodiment, the target maintenance object is a first-level headquarters of a group, with a preset statistical period of 1 month and a sliding time window of the most recent 4 weeks. This first-level headquarters has 5000 active users and manages 6000 devices, operating in a 24 / 7 monitoring mode with an automation coverage rate of 60%. The user-side base and device-side base are calculated: the user-side base is 5000 ÷ 500 = 10, and the device-side base is 6000 ÷ 1000 = 6; therefore, the basic configuration number is taken as 10. The work order fulfillment data from the most recent 4 weeks is then processed. Assuming the target value for the on-time response rate of event work orders is 95% and the actual value is 91%, the event response deviation is (95% - 91%) ÷ 95%. Assuming the target value for the event work order closure rate is 95% and the actual value is 93%, the event closure deviation is (95% - 93%) ÷ 95%. Assuming the target value for the reopening rate is 5% and the actual value is 8%, the event reopening deviation is (8% - 5%) ÷ 5%. The deviation value is calculated using the same method for problematic work orders. After weighting each deviation value according to a preset weight, the risk components for event work orders and problematic work orders are obtained. These are then aggregated based on the number of event work orders, the number of problematic work orders, and their priority distribution, resulting in a risk adjustment factor of 0.28. Since this value is higher than the high-risk threshold of 0.25 obtained through historical sample calibration in this embodiment, it is mapped to a risk correction number of 2.8. Meanwhile, the additional duty roster corresponding to 7×24-hour duty is taken as 50% of the basic configuration number, i.e., 5. The automation reduction number corresponding to 60% automation coverage is 0.6. Therefore, the initial staffing number is 10 + 2.8 + 5 - 0.6 = 17.2, which can be rounded down to 17. Subsequently, constraint calibration is performed: based on the target staffing number of 16 in the previous statistical period, the upper limit of a single adjustment is set at 15%, thus the allowed upper limit is 18.4. After verification of staffing coverage, 17 staff members meet the minimum number required for a shift. However, after verification of performance capability, the predicted on-time response rate with 17 staff members is only 93.6%, lower than the preset requirement of 95%. Therefore, the candidate value is revised to 18. After further verification, the predicted on-time response rate with 18 staff members is 95.4%, and the closed-loop rate is 96.1%, meeting the constraints of skill coverage and staffing fluctuations. Therefore, the target staffing number is set at 18. If the risk adjustment factors for the next month and the month after that are 0.27 and 0.29 respectively, then the risk is considered to be continuously high for two consecutive statistical periods, and a new staffing adjustment will be triggered without exceeding the preset adjustment limit. If the risk is only higher than the high-risk threshold for one month and then falls back to the lag range in the following month, the number of staff will remain unchanged at 18.

[0032] Understandably, a risk adjustment factor is generated using a sliding time window, which is then used to generate the initial staffing levels. A calibration is performed under multiple constraints, including duty coverage, skill coverage, performance compliance, and staffing fluctuations. Finally, a decision on whether to implement dynamic adjustments is made based on continuous statistical cycles, thus forming a two-layer control chain. This embodiment can suppress misjudged staffing increases caused by concentrated weekly failures, temporary release windows, or short-term business activities, reduce frequent back-and-forth changes in staffing levels between adjacent statistical periods, and maintain the continuity of key positions and the stability of scheduling. It avoids overreacting to short-term fluctuations and can promptly release staffing increases when facing continuous performance deterioration, thus possessing both resilience and responsiveness.

[0033] In some embodiments of this application, when generating a risk adjustment factor based on work order performance data, the process includes: determining the event work order risk component and the problem work order risk component within a sliding time window, and aggregating the event work order risk component and the problem work order risk component based on the number of event work orders, the number of problem work orders, and the priority distribution to generate a risk adjustment factor.

[0034] In some embodiments of this application, when determining the risk components of event work orders and problem work orders respectively, the following steps are taken: determining the deviation values ​​of event work orders and problem work orders based on timely response, timely closure, reopening, and backlog, respectively, and weighting the deviation values ​​of event work orders and problem work orders to obtain the risk components of event work orders and problem work orders.

[0035] Specifically, in some embodiments of this application, risk adjustment factors are generated based on work order fulfillment data, primarily used to characterize the fulfillment pressure level of the current maintenance object during the most recent continuous operation period. Unlike methods that directly judge service pressure based solely on a statistical point in time or the number of a certain type of work order, this embodiment divides work order fulfillment data into two parts: event-related work order data and problem-related work order data, which are processed separately. Event-related work orders are mainly used to characterize the immediate response pressure for faults, alarms, or sudden anomalies, while problem-related work orders are mainly used to characterize the pressure of recurring faults, root cause elimination, and long-term closed-loop management. A sliding time window is used to extract work order fulfillment data from the most recent continuous period in terms of time dimension, to reduce the interference of occasional work order spikes on the risk assessment results. Preferably, the sliding time window can be set to 7 days, 14 days, 28 days, or 1 month, and maintain a fixed correspondence with a preset statistical period. For example, in a monthly calculation scenario, the most recent 4 weeks are used as the sliding time window. Work order fulfillment data includes at least the timeliness of response, timeliness of closure, reopening status, backlog status, number of event work orders, number of problem work orders, and priority distribution. Among them, the timeliness of response indicates whether the work order was accepted within the agreed response time limit; the timeliness of closure indicates whether the work order was completed within the agreed closure time limit; the reopening status indicates whether the work order was reopened due to incomplete processing; the backlog status indicates the number of work orders that have not been closed by the end of the sliding time window; and the priority distribution indicates the proportion of work orders with different levels of urgency among all work orders.

[0036] In this embodiment, when determining the risk components of event work orders and problem work orders, corresponding performance indicators are first extracted for each type of work order. Then, the deviation values ​​for event work orders and problem work orders are determined based on the degree of deviation between each indicator and the target value. For indicators where "a larger value indicates a better performance status," such as response timeliness and closure timeliness, the deviation value can be determined as follows: Deviation = (Target value - Actual value) / Target value.

[0037] When the actual value is not lower than the target value, the deviation value is set to 0. For indicators where "a larger value indicates a worse performance status," such as the reopening rate and backlog rate, the deviation value can be determined as follows: Deviation = (Actual value - Target value) / Target value.

[0038] When the actual value is not higher than the target value, the deviation value is set to 0. The purpose of the above processing method is to include only deviations exceeding the risk direction in the calculation, without applying a negative value to indicators that are better than the target value, thereby avoiding a single good value indicator masking other key abnormal indicators. Subsequently, the response deviation value, closed-loop deviation value, reopening deviation value, and backlog deviation value of the event work order are assigned event weights and summed to obtain the event work order risk component. The corresponding deviation values ​​of the problem work orders are assigned problem weights and summed to obtain the problem work order risk component. Correspondingly, the event work order risk component and the problem work order risk component can be expressed as follows: Event work order risk component = event response deviation value × event response weight + event closure deviation value × event closure weight + event reopening deviation value × event reopening weight + event backlog deviation value × event backlog weight.

[0039] Problem work order risk component = Problem response deviation value × Problem response weight + Problem closure deviation value × Problem closure weight + Problem reopening deviation value × Problem reopening weight + Problem backlog deviation value × Problem backlog weight.

[0040] Each weight represents the degree of impact of the corresponding indicator on performance risk. Preferably, the weight of timely response and backlog in event work orders is higher than that of timely closure, to highlight the real-time requirements for handling emergencies. The weight of timely closure and reopening in problem work orders is higher than that of timely response, to highlight the continuous requirements for root cause management and reduction of recurring failures. The sum of all weights is preferably 1 to keep the component results within a uniform dimension. The weights can be derived from historical performance sample analysis results, operational system agreements, or expert-calibrated results. For example, the correlation between abnormal indicators and service default events over the past 12 statistical periods can be statistically analyzed, and indicators with higher correlation can be assigned higher weights. Alternatively, a weight range matching the impact of different priority work orders on business continuity can be set.

[0041] After obtaining the risk components of event work orders and issue work orders, the two are aggregated to generate a risk adjustment factor. Since event work orders and issue work orders typically differ in quantity and business impact, this embodiment does not use a simple averaging method, but instead performs weighted aggregation based on the number of event work orders, the number of issue work orders, and priority distribution. Preferably, the quantity weight can be calculated first based on the number of event work orders and the number of issue work orders, and then the priority adjustment weight can be calculated based on the proportion of high-priority work orders, ultimately generating the risk adjustment factor. Correspondingly, the risk adjustment factor can be expressed as: Risk adjustment factor = event work order risk component × event aggregation weight + problem work order risk component × problem aggregation weight + priority adjustment value.

[0042] The event aggregation weight and issue aggregation weight are used to reflect the relative proportions of the two types of work orders in the total performance pressure, while the priority adjustment value is used to reflect the amplifying effect of high-priority work orders on the overall risk. Preferably, when the proportion of high-priority work orders is higher than a preset proportion threshold, the priority adjustment value increases with the increase of the proportion. The preset proportion threshold can be obtained based on historical statistics, for example, taking the 75th percentile of the proportion of high-priority work orders in the past 12 statistical periods of the same level of maintenance objects as the threshold, so that the threshold is more in line with existing operating patterns. Through this aggregation process, the risk adjustment factor finally obtained is no longer a simple statistical result of the number of work orders, but a comprehensive risk characterization quantity that simultaneously reflects the degree of performance deviation, the difference in work order structure, and the impact of high-priority work orders.

[0043] Specifically, the process involves reading event and issue work order details within the sliding time window, grouping them by work order type and priority, and then extracting the response timestamp, loop closure timestamp, reopening flag, and non-loop closure status flag for each group. These values ​​are then compared with the preset response time limit, loop closure time limit, reopening rate cap, and backlog rate cap in the service level agreement to obtain various deviation values. To ensure data comparability, a minimum sample threshold can be set for cases with significantly insufficient work orders. For example, when the number of a certain type of work order is lower than the preset sample size, the stable value of the previous sliding time window or a set default benchmark value is used for compensation to avoid abnormal fluctuations in components due to insufficient samples. For abnormal extreme values, such as abnormal response times caused by system interruptions or batch imports, abnormal samples exceeding a preset multiple can be removed through extreme value removal before calculating the deviation value. This process does not change the risk itself but rather aims to improve the stability of the risk adjustment factor calculation.

[0044] Specifically, in this embodiment, the sliding time window is set to the most recent 28 days, the number of event work orders is 120, the number of problem work orders is 40, and the proportion of high-priority work orders is 25%. Assuming the target value for event work order response timeliness is 95% and the actual value is 90%, the event response deviation is (95% - 90%) ÷ 95% = 0.0526. Assuming the target value for event work order closure timeliness is 95% and the actual value is 92%, the event closure deviation is (95% - 92%) ÷ 95% = 0.0316. Assuming the target value for event work order reopening rate is 5% and the actual value is 8%, the event reopening deviation is (8% - 5%) ÷ 5% = 0.6. Assuming the target value for event work order backlog rate is 6% and the actual value is 9%, the event backlog deviation is (9% - 6%) ÷ 6% = 0.5. If the event response weight, event closure weight, event reopening weight, and event backlog weight are 0.35, 0.2, 0.2, and 0.25 respectively, then the event work order risk component is 0.0526×0.35 + 0.0316×0.2 + 0.6×0.2 + 0.5×0.25, which is approximately 0.2697. Further, assuming the problem work order response deviation, closure deviation, reopening deviation, and backlog deviation are calculated to be 0.03, 0.08, 0.25, and 0.18 respectively, with corresponding weights of 0.15, 0.35, 0.3, and 0.2, then the problem work order risk component is approximately 0.1445. Further, based on the number of event work orders and the number of problem work orders, the event aggregation weight is determined to be 120÷160=0.75, and the problem aggregation weight is 40÷160=0.25. When the proportion of high-priority work orders (25%) exceeds the preset threshold of 20%, a priority correction value of 0.03 is generated according to the preset mapping relationship corresponding to the excess proportion. Therefore, the risk adjustment factor is 0.2697 × 0.75 + 0.1445 × 0.25 + 0.03, with a calculated result of approximately 0.2684. This result indicates that the maintenance objects within the current sliding time window are in a state of high performance risk, and can be used for preliminary data generation and dynamic adjustment judgment.

[0045] Understandably, by modeling event work orders and problem work orders separately, and further introducing priority distribution to correct the convergence results, it is possible to identify two different types of performance risks—"increased pressure for immediate handling" and "increased pressure for root cause management"—when the total number of work orders is similar. This avoids the risk masking effect caused by the existing unified calculation based on total or average performance indicators. This embodiment improves the resolution and foresight of the risk identification stage. Even if the total number of work orders does not increase significantly, as long as the proportion of high-priority event work orders increases or problem work orders are reopened or the backlog worsens, the risk adjustment factor can be amplified in a timely manner, allowing the preparation and calibration to perceive potential performance pressure in advance. This not only improves the accuracy of risk characterization but also enhances the foresight and targeting of personnel allocation adjustments.

[0046] In some embodiments of this application, when constraining and calibrating the initial staffing number to obtain the target staffing number, the process includes: taking the initial staffing number as the initial value, sequentially verifying the duty coverage constraint, skill coverage constraint, performance constraint, and staffing fluctuation constraint, correcting the initial staffing number that does not meet the constraints, until the target staffing number that meets all constraints is obtained.

[0047] Specifically, in some embodiments of this application, a constraint calibration is performed on the preliminary staffing number to obtain the target staffing number. This is mainly used to address the issue that while the preliminary staffing number can reflect the current operational pressure, it may not be directly converted into an executable personnel configuration result. The preliminary staffing number is usually generated by combining the basic configuration number, risk adjustment results, additional duty personnel, and automated staff reduction numbers. It is essentially a calculation value oriented towards the overall load and may have issues such as non-integer values, incomplete skill distribution, insufficient shift coverage, or rapid changes compared to the previous statistical period. Therefore, this embodiment sets up a constraint calibration process after the preliminary staffing number is generated. The preliminary staffing number is used as the initial value to form candidate staffing numbers. Then, the candidate staffing numbers are sequentially subjected to duty coverage constraint verification, skill coverage constraint verification, performance constraint verification, and staffing fluctuation constraint verification. Candidate staffing numbers that do not meet the constraints are corrected until the target staffing number that meets all constraints is obtained. Preferably, the candidate staffing number is first integerized. The integerization method can be rounding up or rounding to the nearest integer. In scenarios with high performance pressure or heavy duty personnel, rounding up is preferred to reduce the risk of insufficient configuration due to rounding.

[0048] In this embodiment, the duty coverage constraint is used to ensure that the target maintenance object has a minimum on-duty support capability in each time period under the established duty mode. The duty coverage constraint is related to at least the number of duty shifts, the minimum number of personnel on duty for each shift, the rotation coverage coefficient, and the number of emergency backup personnel. Among them, the rotation coverage coefficient is used to reflect the reduction effect of factors such as leave, rest, training, sick leave, personal leave, and temporary replacements on the nominal staffing. Accordingly, the minimum number of on-duty personnel can be determined by the following formula: Minimum number of on-duty personnel = Sum of minimum on-duty personnel for each shift × Rotation coverage coefficient + Number of emergency backup personnel. The rotation coverage coefficient can be obtained by calibration based on the shift system and historical attendance data. For example, in a 7×24-hour three-shift scenario, a value between 4.0 and 5.0 can be taken, and in a weekday day shift scenario, a value between 1.1 and 1.5 can be taken. If the number of candidate staffing is less than the minimum number of on-duty personnel, it is determined that the duty coverage constraint is not met, and the number of candidate staffing is adjusted to at least an integer value not lower than the minimum number of on-duty personnel.

[0049] Skill coverage constraints ensure that the target maintenance personnel possess the critical skill set required to complete maintenance tasks while meeting the headcount requirements. Skill coverage constraints are related to at least the skill domain division results, the minimum headcount for each skill domain, and the personnel's composite skill coverage capabilities. Skill domains can be categorized based on the target maintenance personnel's responsibilities, such as infrastructure maintenance, system maintenance, network maintenance, database maintenance, application support, work order coordination, or on-site handling. The minimum headcount for each skill domain can be determined based on the service catalog, historical fault type distribution, and organizational job requirements. Preferably, when a person possesses capabilities in multiple skill domains and reaches a preset skill proficiency threshold, they can be counted as a composite skill personnel in multiple skill domain coverage calculations; however, to prevent excessive sharing, a single person can be counted in a maximum of two skill domains. If the job structure corresponding to the candidate headcount cannot meet the minimum headcount requirements for all skill domains, the skill coverage constraint is deemed not met, and the candidate headcount is adjusted according to the headcount gap corresponding to the missing skill domains.

[0050] Performance constraints are used to ensure that the service capacity corresponding to the candidate staffing level can meet the performance targets within the current statistical period. Performance constraints no longer rely on the calculation results of the aforementioned risk adjustment factors, but instead verify the results against the target staffing level. Specifically, the average processing capacity per person per cycle, the average processing time for work orders of different priorities, and the peak-hour load amplification factor can be obtained first based on historical sample statistics. Then, the predicted processing capacity under the candidate staffing level can be calculated accordingly. The predicted processing capacity can be determined by the following formula: Predicted processing capacity = Candidate staffing level × Effective working hours per person per cycle ÷ Average processing time per work order. Simultaneously, the demand capacity is obtained based on the predicted work order volume and priority distribution for the current statistical period. The average processing time per work order is preferably obtained by weighting the average processing time of work orders of different priorities with their corresponding proportions. The demand capacity is preferably determined by multiplying the predicted work order volume by the peak-hour load amplification factor. The peak-hour load amplification factor can be calibrated based on the ratio of the daily peak work order volume to the daily average work order volume over the past 12 statistical periods, preferably between 1.1 and 1.5. The on-time response rate threshold and on-time closure rate threshold are preferably derived from the service level agreement, operation and maintenance assessment standards, or a stable compliance range over the past 12 statistical periods. When the predicted processing capacity is not lower than the demand capacity, and the on-time response rate and on-time closure rate calculated from the candidate staffing numbers are not lower than the preset fulfillment thresholds, the fulfillment constraint is deemed met. The fulfillment thresholds are preferably derived from the service level agreement, operation and maintenance assessment standards, or a stable compliance range over the past 12 statistical periods; for example, the on-time response rate threshold can be 95%, and the on-time closure rate threshold can be 90% or 95%. If the candidate staffing numbers cannot meet the above requirements, adjustments are made incrementally according to a preset adjustment step size. Preferably, the adjustment step size is 1 person to ensure a smooth adjustment process and ease of implementation.

[0051] Staffing fluctuation constraints are used to control the magnitude of change in the target staffing number relative to the target staffing number in the previous preset statistical period, avoiding frequent and significant changes in staffing results due to short-term load fluctuations. Staffing fluctuation constraints include at least an upper limit for the increase and an upper limit for the decrease, which can be the same or different. Preferably, the upper limit for the increase and the upper limit for the decrease are determined as a certain percentage of the target staffing number in the previous preset statistical period, for example, a range of 10% to 20%. This percentage can be determined based on the organization's manpower adjustment pace, recruitment cycle, difficulty of job replacement, and historical staffing fluctuation statistics. After the candidate staffing number meets the duty coverage constraint, skill coverage constraint, and performance constraint, it is then compared with the fluctuation upper limit. If multiple candidate staffing numbers meet all constraints simultaneously, the candidate staffing number with the smallest difference from the initial staffing number and the fewest personnel is preferably selected as the target staffing number, to balance performance capability and resource conservation.

[0052] Specifically, in this embodiment, the initial staffing number for a target maintenance object is 17.2, which is rounded up to obtain a candidate staffing number of 17. This target maintenance object operates on a 7×24-hour shift model, with a minimum of 2 staff members on duty per shift and 1 emergency backup staff member. The rotation coverage coefficient is 4.2, so the minimum number of staff on duty = 2×4.2 + 1 = 9.4, which is rounded up to 10. Therefore, the candidate staffing number of 17 satisfies the duty coverage constraint. Furthermore, the target maintenance object requires 2 infrastructure maintenance personnel, 2 network maintenance personnel, 3 application support personnel, 1 database maintenance personnel, and 2 work order coordinators, for a total minimum skill coverage of 10. Two of these personnel possess both infrastructure and network maintenance capabilities, so the candidate staffing number of 17 satisfies the skill coverage constraint. Subsequently, a performance constraint verification is performed. Assuming an effective working time of 140 hours per person per unit cycle and an average processing time of 9 hours per work order, the predicted processing capacity = 17×140÷9, approximately 264 cases. If the weighted demand capacity predicted for the current statistical period is 278 units, then the candidate staffing number 17 does not meet the fulfillment constraint. Therefore, it is adjusted to 18 with an adjustment step of 1 person. When 18 is re-verified, the predicted processing capacity increases to 280 units, which meets the fulfillment constraint. Comparing the staffing fluctuation constraint, assuming the target staffing number for the previous preset statistical period is 16, and the upper limit of the increase is 15%, then the maximum allowable target staffing number is 18.4. 18 is acceptable when rounded down or configured as an integer. Therefore, 18 also meets the staffing fluctuation constraint, and 18 is finally determined as the target staffing number.

[0053] Understandably, this approach transforms a total estimated number into a target staffing level that can be scheduled, assigned to specific roles, fulfilled, and smoothly implemented. This embodiment improves the executability of the results. Even if the initial staffing level already reflects changes in risk and workload, it can further eliminate hidden defects such as "sufficient staff but inability to schedule shifts," "meeting the total target but lacking key skills," and "excessive monthly fluctuations making implementation difficult." In particular, by placing fulfillment constraints after skill coverage constraints and before staffing fluctuation constraints, this embodiment can first ensure the bottom line of operational capabilities and then control the adjustment range. This achieves a balance between fulfillment capabilities and organizational stability without significantly increasing staffing redundancy. This balancing effect brought about by the order of constraints is usually difficult to obtain when relying solely on total amount calculations.

[0054] In some embodiments of this application, the determination of continuous statistical periods includes: if the risk adjustment factor is consistently higher than the high-risk threshold for at least two consecutive statistical periods, it is determined to be a sustained high-risk situation; if the risk adjustment factor is consistently lower than the low-risk threshold, it is determined to be a sustained low-risk situation; and if the risk adjustment factor is in the hysteresis range, it is determined to be a state of sustained volatility.

[0055] In some embodiments of this application, the target number of staffing is increased, decreased, or kept unchanged depending on whether the risk is high, low, or fluctuating. The dynamically adjusted target number of staffing is then used as the reference number for the next preset statistical period.

[0056] Specifically, in some embodiments of this application, after calculating the target number of entries for the current preset statistical period, a continuous statistical period determination and dynamic adjustment process is further performed to determine whether the current risk status is a short-term fluctuation or a persistent trend that requires adjustment. The continuous statistical period determination result is not a simple comparison of the risk adjustment factor at a single point in time, but a status result obtained after judging the continuity of the interval points of the risk adjustment factor within at least two adjacent preset statistical periods. The continuous statistical period determination result can be understood as the output status. The current preset statistical period first determines the candidate status based on the relationship between the risk adjustment factor and the high-risk threshold, low-risk threshold, and hysteresis interval, and then combines the candidate status of the previous preset statistical period to obtain the output status. Only when both the candidate status of the current preset statistical period and the previous preset statistical period are high-risk, the output is "continuously high-risk". Only when both the candidate status of the current preset statistical period and the previous preset statistical period are low-risk, the output is "continuously low-risk". When the candidate status of the current preset statistical period is within the hysteresis interval, the output is "fluctuation-maintained status". Specifically, for each preset statistical period, the corresponding risk adjustment factor, target compilation number, and adjustment execution record are saved, forming a periodic state sequence in chronological order. Then, the risk adjustment factor for the current preset statistical period is compared with the high-risk threshold, low-risk threshold, and hysteresis interval to determine the candidate state corresponding to the current period. If both the current and previous period's candidate states are high-risk, a "continuously high-risk" state is output. If both the current and previous period's candidate states are low-risk, a "continuously low-risk" state is output. If the current period's risk adjustment factor is within the hysteresis interval, a "fluctuation-maintained" state is output. This method combines "threshold comparison" with "time continuity confirmation," preventing occasional anomalies within a single statistical period from directly triggering compilation adjustments.

[0057] In this embodiment, the high-risk threshold, low-risk threshold, and hysteresis interval are preferably calibrated based on historical operation and maintenance sample data. Specifically, the risk adjustment factor of the same-level target operation and maintenance object within 12 or 24 consecutive preset statistical periods can be correlated with the corresponding performance compliance rate, overdue work order ratio, and key fault impact results to obtain the relationship between the risk adjustment factor and the probability of performance instability. When the risk adjustment factor rises to a certain value and the probability of performance instability increases significantly, this value is determined as the high-risk threshold. When the risk adjustment factor falls below a certain value and the performance indicators remain stable and compliant for a long period, this value is determined as the low-risk threshold. The interval between the two is defined as the hysteresis interval. The purpose of setting the hysteresis interval is to retain a buffer area between the high-risk threshold and the low-risk threshold that does not directly trigger increases or decreases in configuration, thereby suppressing frequent switching caused by small fluctuations near the boundary. Preferably, the high-risk threshold and the low-risk threshold can be jointly determined based on historical data distribution, service level agreement requirements, and the organization's acceptable fluctuation range. When the target operation and maintenance object undergoes significant business adjustments, service catalog changes, or automation coverage level changes, threshold calibration can be re-performed to maintain the consistency of the threshold with the actual operation and maintenance environment.

[0058] After determining the continuous statistical period's assessment result, corresponding staffing adjustments are executed based on whether the risk level remains high, low, or fluctuates. Specifically, when the continuous statistical period's assessment result is high risk, the target staffing number for the current preset statistical period is increased. When the continuous statistical period's assessment result is low risk, the target staffing number for the current preset statistical period is decreased. When the continuous statistical period's assessment result is fluctuating, the target staffing number for the current preset statistical period remains unchanged. To ensure the feasibility of the adjustments, the increase and decrease are not unlimited but are executed according to a preset adjustment step size or preset adjustment ratio. The adjusted target staffing number can be determined as follows: the dynamically adjusted target staffing number equals the current target staffing number plus the increase, or equals the current target staffing number minus the decrease. The increase and decrease are jointly determined by the continuous statistical period's assessment result, the deviation of the current risk adjustment factor from the threshold, and the upper limit of staffing adjustments. When the deviation is small, a fixed step size of one person is preferred. When the deviation is large and persists for multiple periods, adjustments can also be made proportionally, provided that the upper limit of staffing adjustments is not exceeded. After dynamic adjustment is completed, the dynamically adjusted target number is used as the reference number for the next preset statistical period. It is used for the generation of the initial number of the next period, constraint calibration or adjustment range control, thus forming a closed-loop processing link of "current judgment, current adjustment, and inheritance in the next period".

[0059] It should be noted that the compilation reference number is not a fixed value that replaces the entire calculation process of the next preset statistical period, but rather serves as a benchmark value for the initial compilation calculation and compilation fluctuation control in the next preset statistical period. On the one hand, the next preset statistical period can still regenerate a new initial compilation number based on new business load data, work order fulfillment data, duty demand data, and automation coverage data. On the other hand, the dynamically adjusted target compilation number, as a compilation reference number, can be used to limit the increase or decrease in the next period, avoiding drastic jumps between adjacent weeks. Correspondingly, even if a period experiences sustained high risk and completes the allocation, subsequent periods will not exist independently of the new operating data, but will continue to be dynamically corrected based on the inherited compilation reference number, thereby ensuring that the compilation adjustment is both continuous and responsive to the actual operating status.

[0060] Specifically, in this embodiment, the target maintenance object is statistically analyzed monthly. The risk adjustment factor for the current month is 0.31, and the risk adjustment factor for the previous month is 0.29. The high-risk threshold obtained through historical sample calibration is 0.27, the low-risk threshold is 0.18, and the hysteresis interval is 0.18 to 0.27. Since the risk adjustment factors for both the current and previous months are higher than the high-risk threshold, the continuous statistical period determination result is determined to be continuously high-risk. If the target staffing number obtained through constraint calibration for the current month is 18, the preset adjustment step size is 1, and the maximum adjustment ratio in a single instance is 15% of the target staffing number for the previous period, and the target staffing number for the previous period was 17, then the maximum allowable number of people to be adjusted this time is 2.55, rounded to 2. Since adding 1 person at a fixed step size is sufficient to meet the adjustment strategy of this embodiment, the dynamically adjusted target staffing number is determined to be 19, and 19 is used as the staffing reference number for the next month. For example, if the risk adjustment factor in a subsequent month falls back to 0.23, it is considered to be within the lag range, and that month is classified as maintaining volatility, with the target allocation of 19 remaining unchanged from the previous month. Only when the risk adjustment factor for two consecutive statistical periods is below 0.18 will a reduction in allocation be triggered. This process demonstrates that this embodiment does not adjust based solely on a single risk outcome for the current month, but rather continuously confirms the risk trend before implementing adjustments, thereby improving the reliability of the adjustment results.

[0061] Understandably, this embodiment reduces the repeated fluctuations near the threshold boundary. Even if the risk adjustment factor fluctuates around the threshold between adjacent weeks, it will not immediately lead to frequent switching in the direction of staffing increases or decreases. Simultaneously, when the risk status continuously deteriorates or improves, the adjustment effect can be continued in subsequent cycles through the reference number inheritance mechanism, avoiding the need to reassess from scratch in each cycle. Therefore, this embodiment further enhances the continuity of staffing evolution and the predictability of organizational execution while maintaining dynamism, transforming staffing management from "periodic calculation" to "continuous adjustment with inertial control."

[0062] In some embodiments of this application, when increasing the target staffing level based on persistent high risk, the target staffing level is increased according to a preset adjustment step size. When decreasing the target staffing level based on persistent low risk, the target staffing level is decreased according to a preset adjustment step size.

[0063] In some embodiments of this application, when the risk adjustment factor changes from a state of sustained high risk or sustained low risk to a state of sustained volatility, the target compilation number of the previous preset statistical period is continued.

[0064] Specifically, in some embodiments of this application, after the preliminary processing has obtained the continuous statistical cycle determination results of "continuous high risk," "continuous low risk," or "fluctuation-maintained state," a further discrete step-by-step adjustment of staffing is performed. The preset adjustment step size is used to limit the minimum adjustment unit of the target staffing number within a single preset statistical cycle, essentially corresponding to the minimum executable granularity of personnel allocation in organizational implementation. Since maintenance positions are usually implemented based on natural persons, shift personnel, or job units, the preset adjustment step size is preferably an integer, specifically 1 person, 2 people, or determined according to the personnel unit corresponding to the job group. For target maintenance objects with a smaller staffing scale and more detailed job division, 1 person is preferred. For target maintenance objects with a larger staffing scale and replaceable positions, it can also be determined as an integer value not exceeding a preset proportion of the current target staffing number. The source of the preset adjustment step size can be comprehensively determined based on historical staffing adjustment records, job replacement cycles, the smallest implementation unit of recruitment or adjustment, and the smallest change granularity of scheduling. Its purpose is to avoid dynamic adjustment results that are computationally valid but cannot be implemented in organizational practice.

[0065] In this embodiment, when increasing the target staffing level based on persistent high risk, the target staffing level is increased by a preset adjustment step size. When decreasing the target staffing level based on persistent low risk, the target staffing level is decreased by a preset adjustment step size. Accordingly, the dynamically adjusted target staffing level can be expressed as follows: when the determination result is persistent high risk, the dynamically adjusted target staffing level is equal to the current target staffing level plus the preset adjustment step size. When the determination result is persistent low risk, the dynamically adjusted target staffing level is equal to the current target staffing level minus the preset adjustment step size. It should be noted that the current target staffing level refers to the target staffing level obtained after constraint calibration within the current preset statistical period, not the preliminary staffing level obtained from previous calculations. Furthermore, to prevent the reduction from compromising existing executability, when performing the reduction, the reduction result should be compared with the lowest executable staffing lower limit obtained from constraint calibration within the current preset statistical period. When the result after reduction is lower than the lowest executable staffing lower limit, the dynamically adjusted target staffing level takes the lowest executable staffing lower limit. The reason for this approach is that a sustained low risk only indicates a relative decrease in operational pressure, and does not necessarily mean that the minimum requirements for on-duty coverage, skill coverage, and contract fulfillment can be exceeded. Therefore, the reduction in staffing must still be constrained by the current executable boundaries. The minimum executable staffing limit refers to the minimum integer staffing number required to simultaneously meet the constraints of on-duty coverage, skill coverage, and contract fulfillment within the current preset statistical period.

[0066] In some embodiments of this application, when the risk adjustment factor transitions from a persistently high-risk or persistently low-risk state to a volatility-maintaining state, the target allocation number from the previous preset statistical period is continued. Here, "continued" does not simply mean using the historical value before adjustment, but rather using the target allocation number that has been actually determined and is prepared for execution in the previous preset statistical period. If the previous preset statistical period has already completed an increase in allocation based on persistently high risk, the target allocation number after the increase will continue to be maintained when entering the volatility-maintaining state. If the previous preset statistical period has already completed a decrease in allocation based on persistently low risk, the target allocation number after the decrease will continue to be maintained when entering the volatility-maintaining state. The volatility-maintaining state indicates that the risk adjustment factor has left the persistently high-risk or persistently low-risk range, but has not yet formed a reverse persistent trend. If, at this time, the allocation is immediately reduced or increased in the opposite direction according to the current single-period result, it is easy to generate back-and-forth oscillations near the risk boundary, such as "increase-retreat-re-increase" or "decrease-replenishment-re-decrease." Therefore, in this embodiment, when the state changes from a sustained high-risk or sustained low-risk state to a fluctuating state, no compensation adjustment in the opposite direction is performed. Instead, the target compilation number of the previous preset statistical period is directly used as the continuation value of the current period and continues to serve as the compilation reference basis for subsequent periods. If subsequent consecutive statistical periods again form a sustained high-risk or sustained low-risk state, a new increase or decrease in allocation is performed according to the preset adjustment step size.

[0067] The system reads the results of continuous statistical cycles and the current target allocation. Based on a preset adjustment step size, it generates an increase or decrease value and outputs the dynamically adjusted target allocation candidate value. When the state transitions from a sustained high-risk or sustained low-risk state to a fluctuating, stable state, it blocks reverse adjustment commands within the current cycle and copies the target allocation from the previous preset statistical cycle as the effective value for the current cycle. Preferably, it can also record the state transition sequence of the most recent multiple preset statistical cycles to determine whether there is a repeated swing between "high risk—stable—high risk" or "low risk—stable—low risk". When a repeated swing exists, the original preset adjustment step size can be maintained unchanged without further increasing the adjustment amplitude to ensure the consistency of the dynamic adjustment logic.

[0068] Specifically, in this embodiment, suppose the target staffing number of a certain target maintenance object is 18 after constraint calibration in the first statistical period, and the continuous statistical period judgment result is a persistent high risk, with a preset adjustment step size of 1. Then, the dynamically adjusted target staffing number is 18 + 1 = 19. In the second statistical period, if the risk adjustment factor still meets the persistent high risk condition, and the current target staffing number after constraint calibration is 19, then the dynamically adjusted target staffing number is 19 + 1 = 20. Subsequently, in the third statistical period, if the risk adjustment factor changes from a persistent high risk state to a fluctuating state, the target staffing number is not reverted to 19, but directly continues the target staffing number of 20 formed in the previous preset statistical period. Conversely, if the current target staffing number of another target maintenance object is 16 in a certain statistical period, the continuous statistical period judgment result is a persistent low risk, the preset adjustment step size is also 1, and the current minimum executable staffing limit is 15, then the dynamically adjusted target staffing number is 16 - 1 = 15. If the risk adjustment factor in the next statistical period changes from a sustained low-risk state to a fluctuating state, the target compilation number of 15 from the previous preset statistical period will continue, instead of immediately rising to 16. As can be seen from the above process, the dynamic adjustment in this embodiment is not a continuous variable adjustment, but an executable adjustment based on discrete step length and state transition rules.

[0069] Understandably, this embodiment suppresses over-correction during staffing adjustments. Specifically, when the risk status has just shifted from a sustained high-risk or sustained low-risk state, it does not immediately implement a reverse swing. Instead, it maintains the target staffing level from the previous preset statistical period, preserving a stable window for job handover, shift rescheduling, and knowledge transfer. The resulting effect is not merely "reducing the number of adjustments," but more importantly, maintaining a relatively stable manpower support level during periods of temporary risk reduction but before service pressure has completely dissipated. This avoids the phenomenon of immediately downsizing after a concentrated period of fault handling, followed by a resurgence of staffing due to backlog. This embodiment, through a combination of "fixed step size adjustment + continued execution value during status transition," evolves staffing management from a simple threshold response into a discrete control process with execution buffer capabilities, creating an anti-overshoot effect at the organizational implementation level.

[0070] Based on another preferred embodiment described above, see [link to preferred embodiment]. Figure 2 As shown, this embodiment provides a maintenance personnel allocation processing system for applying the above-described maintenance personnel allocation processing method, including: The data acquisition unit is used to acquire work order fulfillment data, business load data, duty requirement data, and automation coverage data of the target maintenance object within a preset statistical period.

[0071] The processing unit is used to determine the basic configuration number based on business load data, generate risk adjustment factors based on work order fulfillment data, determine the additional on-duty personnel based on on-duty requirement data, determine the automation reduction number based on automation coverage data, and generate a preliminary staffing number based on the basic configuration number, risk adjustment factors, additional on-duty personnel, and automation reduction number.

[0072] The calibration unit is used to perform constraint calibration on the initial staffing number to obtain the target staffing number. The constraints include duty coverage constraints, skill coverage constraints, performance constraints, and staffing fluctuation constraints.

[0073] The adjustment unit is used to dynamically adjust the target number based on the correspondence between the risk adjustment factor and the preset risk threshold range, and in combination with the judgment results of the continuous statistical period.

[0074] On the other hand, this application also provides a computer-readable storage medium in which the electronic device can perform the above-described method when the computer-executable instructions stored in the computer-readable storage medium are executed by the processor of the electronic device.

[0075] It is understandable that the above-mentioned methods for allocating maintenance personnel, systems, and storage media have the same beneficial effects, and will not be elaborated further here.

[0076] 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 method for processing the allocation ratio of operation and maintenance personnel, characterized in that, include: Acquire work order fulfillment data, business load data, duty requirement data, and automation coverage data of the target maintenance object within a preset statistical period; The basic configuration number is determined based on the business load data, and a risk adjustment factor is generated based on the work order fulfillment data. The additional duty personnel are determined based on the duty requirement data, the automated staff reduction is determined based on the automated coverage data, and a preliminary staffing number is generated based on the basic configuration number, the risk adjustment factor, the additional duty personnel, and the automated staff reduction. The target staffing number is obtained by constraining and calibrating the initial staffing number. The constraints include duty coverage constraints, skill coverage constraints, performance constraints, and staffing fluctuation constraints. Based on the correspondence between the risk adjustment factor and the preset risk threshold range, and combined with the judgment results of continuous statistical periods, the target number of compilations is dynamically adjusted. When generating risk adjustment factors based on the work order fulfillment data, the following are included: Within a sliding time window, the risk components of event work orders and problem work orders are determined respectively, and the risk components of event work orders and problem work orders are aggregated according to the number of event work orders, the number of problem work orders and the priority distribution to generate the risk adjustment factor; When determining the risk components of event work orders and problem work orders respectively, the following should be included: The deviation values ​​of event work orders and problem work orders are determined based on the timeliness of response, timeliness of closure, reopening, and backlog, respectively. The deviation values ​​of event work orders and problem work orders are then weighted to obtain the risk components of event work orders and problem work orders. The continuous statistical period determination results include: if the risk adjustment factor is continuously higher than the high risk threshold for at least two consecutive statistical periods, it is determined to be continuously high risk; if the risk adjustment factor is continuously lower than the low risk threshold, it is determined to be continuously low risk; if the risk adjustment factor is in the hysteresis interval, it is determined to be in a state of sustained volatility. When the risk adjustment factor changes from a state of sustained high risk or sustained low risk to a state of sustained volatility, the target number of the previous preset statistical period is continued.

2. The method for handling the allocation of maintenance personnel according to claim 1, characterized in that, When constraining and calibrating the initial staffing number to obtain the target staffing number, the following steps are included: Using the initial staffing number as the initial value, the duty coverage constraint, the skill coverage constraint, the performance constraint, and the staffing fluctuation constraint are checked in sequence. The initial staffing number that does not meet the constraints is corrected until the target staffing number that meets all constraints is obtained.

3. The method for handling the allocation of maintenance personnel according to claim 1, characterized in that, Based on the state of sustained high risk, sustained low risk, or continued fluctuation, the target number of staffing will be increased, decreased, or kept unchanged, and the dynamically adjusted target number of staffing will be used as the reference number for the next preset statistical period.

4. The method for handling the allocation of maintenance personnel according to claim 3, characterized in that, When increasing the target staffing level based on the sustained high risk, the target staffing level is increased according to a preset adjustment step size; when decreasing the target staffing level based on the sustained low risk, the target staffing level is decreased according to the preset adjustment step size.

5. A system for processing the allocation of maintenance personnel, used to apply the method for processing the allocation of maintenance personnel as described in any one of claims 1-4, characterized in that, include: The data acquisition unit is used to acquire work order fulfillment data, business load data, duty requirement data and automation coverage data of the target maintenance object within a preset statistical period. The processing unit is used to determine the basic configuration number based on the business load data and generate a risk adjustment factor based on the work order fulfillment data. The additional duty personnel are determined based on the duty requirement data, the automated staff reduction is determined based on the automated coverage data, and a preliminary staffing number is generated based on the basic configuration number, the risk adjustment factor, the additional duty personnel, and the automated staff reduction. The calibration unit is used to perform constraint calibration on the preliminary staffing number to obtain the target staffing number. The constraints include duty coverage constraints, skill coverage constraints, performance constraints, and staffing fluctuation constraints. The adjustment unit is used to dynamically adjust the target number of compilations based on the correspondence between the risk adjustment factor and the preset risk threshold range, and in combination with the judgment results of the continuous statistical period.

6. A computer-readable storage medium, characterized in that, When the computer-executable instructions stored in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device is capable of performing the method as described in any one of claims 1 to 4.

Images