Work principal matching system and matching method based on personnel capability matrix

Abstract

The invention relates to the technical field of nuclear power operation and maintenance management, in particular to a work principal matching system and method based on a personnel capability matrix, and the system comprises a database, a capability matrix construction module, a work order feature extraction module and a work order matching module. The database comprises work order historical data, maintenance principal data and historical task data; the capability matrix construction module performs four-dimensional capability index quantification on the maintenance principal by reading the information of the database and constructs a decision matrix; the work order feature extraction module performs feature parameter analysis on the work order and establishes a work order feature model; and the work order matching module reads the information of the database, the decision matrix of the capability matrix construction module and the work order feature model of the work order feature extraction module, and matches the maintenance principal and the work order. According to the system, the decision matrix is constructed through the capability matrix construction module, and the work order matching module matches the work order and the person in charge, so that the work order matching efficiency is improved, and the matching accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of nuclear power operation and maintenance management, and particularly to a work leader matching system and method based on a personnel ability matrix. Background Art

[0002] Nuclear power maintenance work orders are key tools for nuclear power plant maintenance management, used to record detailed information about maintenance tasks, including equipment numbers, maintenance content, required tools, safety measures, etc. With the development of nuclear power technology, the management methods of work orders have also evolved continuously. In the early days, it mainly relied on paper records, which had problems such as slow information transmission and easy errors. Nowadays, with the introduction of digital technology, more and more nuclear power enterprises have begun to adopt electronic work order systems, realizing the rapid generation, approval, execution, and archiving of work orders through information means, greatly improving the maintenance efficiency and the refinement level of management.

[0003] In the prior art, the allocation of nuclear power maintenance work orders relies on manual experience, and problems such as mismatches between skills and tasks, overloading of responsible persons, and delays in urgent tasks are likely to occur. The matching algorithms of current nuclear power maintenance work order allocation systems are difficult to adapt to the multi-dimensional dynamic requirements of nuclear power maintenance scenarios, lacking a dynamic adjustment mechanism and a dynamic update mechanism for the responsible person's ability matrix and a feedback optimization mechanism for task execution results, and there is also a problem of a single ability evaluation dimension, ignoring key factors such as actual experience, task complexity, and dynamic load. Summary of the Invention

[0004] The present invention provides a work leader matching method based on a personnel ability matrix to solve the problems in the prior art that work order allocation relies on manual experience, has a single ability evaluation dimension, and lacks a feedback mechanism.

[0005] The technical solution of the present invention is as follows: The present invention proposes a work leader matching system based on a personnel ability matrix. The system includes a database, an ability matrix construction module, a work order feature extraction module, and a work order matching module. The database includes work order historical data, maintenance responsible person data, and historical task data; the ability matrix construction module quantifies four-dimensional ability indicators for maintenance responsible persons by reading the information in the database and constructs a decision matrix; the work order feature extraction module analyzes the characteristic parameters of the work order and establishes a work order feature model; the work order matching module reads the information in the database, the decision matrix of the ability matrix construction module, and the work order feature model of the work order feature extraction module to match the maintenance responsible person and the work order.

[0006] In some embodiments, the database includes a work order database, a personnel capacity archive, and a historical task database. The work order database stores the historical data of nuclear power maintenance work orders. The work order database performs natural language processing on the unstructured text of nuclear power maintenance work orders, extracts keywords, and maps them to standardized tags. The personnel capacity archive records the qualification certificates of maintenance responsible persons, the maximum task load per person, the current task volume, the historical task completion rate, and violation records. The historical task database stores the execution result data of each work order.

[0007] In some embodiments, the ability matrix construction module quantifies the four-dimensional ability indicators of the maintenance responsible person and constructs a decision matrix, which specifically includes: The ability matrix construction module calculates the skill score according to the qualification certificate level in the personnel capacity archive, as shown in formula (1): (1) Among them, the weight of the certificate level is specifically: the primary certificate level is 0.3, the intermediate certificate level is 0.6, and the advanced certificate level is 1.0; the weight of the skill type is specifically: the core skill is 0.7, and the auxiliary skill is 0.3.

[0008] Based on the execution result data of each work order in the historical task database, the ability matrix construction module classifies and statistically calculates the success rate of the maintenance responsible person according to the equipment type and failure type, and calculates the experience value, as shown in formula (2): The ability matrix construction module collects the current task volume and the estimated available working hours of the maintenance responsible person, calculates the load factor, and the calculation formula of the load factor is as shown in formula (3): The ability matrix construction module reads the violation records in the personnel capacity archive, calculates the safety performance, and the safety performance is specifically as shown in formula (4): Among them, the level coefficient is specifically: a minor violation is 1, a general violation is 3, and a serious violation is 10; after obtaining the four-dimensional ability indicators of the maintenance responsible person, the ability matrix construction module constructs a decision matrix, where each row of the decision matrix represents a maintenance responsible person, and each column is the value of the four-dimensional ability indicator.

[0009] In some embodiments, the ability matrix construction module calculates the entropy value of the ability indicator according to the decision matrix and determines the ability indicator weight, and calculates the ability priority of each responsible person according to the ability indicator weight and the TOPSIS method; the calculation formula of the ability indicator entropy value is specifically formula (5): Among them, e j is the entropy value of the jth ability indicator; m is the number of responsible persons, pij It is the ratio of the value of the i-th person in charge on the j-th ability index to the total value of all persons in charge of this index; The calculation formula for the ability index weight is specifically Formula (6): where, w j is the calculated weight of the j-th ability index; e j is the entropy value of the j-th ability index, and n is the number of ability indexes.

[0010] In some embodiments, the work order feature extraction module reads the information of the work order and establishes a work order feature model from three aspects: complexity, urgency, and safety risk. The specific calculation of the complexity by the work order feature extraction module is as shown in Formula (7): (7) Among them, the skill level requirements are specifically that the primary skill level requirement is 0.3, the intermediate skill level requirement is 0.6, and the advanced skill level requirement is 1.0; the skill type weights are specifically that the core skill = 0.7 and the auxiliary skill = 0.3; The work order feature extraction module determines according to the work classification, and the work classification is determined by the work characteristic parameters of the work order. The work levels are divided into levels 0, 1, 2, 3, and 4. Among them, the work order urgency corresponding to the work level 0 is 1, the work order urgency corresponding to the work level 1 is 0.8, the work order urgency corresponding to the work level 2 is 0.6, the work order urgency corresponding to the work level 3 is 0.4, and the work order urgency corresponding to the work level 4 is 0.2.

[0011] The work order feature extraction module quantifies the work order safety risk, and the safety risk is assigned according to the radiation level and the operation specification level, specifically as shown in Formula (8): (8) Among them, the radiation level coefficient is determined according to the radiation level of the operation area and is obtained from the characteristic parameters of the work order. According to the general standards of nuclear power plants, the radiation level coefficients are divided into the red zone, orange zone, yellow zone, green zone, and no-dose zone. The radiation level coefficient of the red zone is 1, the radiation level coefficient of the orange zone is 0.75, the radiation level coefficient of the yellow zone is 0.5, the radiation level coefficient of the green zone is 0.25, and the radiation level coefficient of the no-dose zone is 0; the operation standardization coefficient depends on the human factors prevention level of the work order and is obtained from the characteristic parameters of the work order.

[0012] In some embodiments, for a single work order, the work order matching module screens out the maintenance person in charge who meets the time window and skill requirements of the work order according to the work order requirements. Using a multi-objective optimization algorithm, the matching degree between each work order and the maintenance person in charge meeting the conditions of the work order is calculated. Taking maximizing the global skill matching degree, maximizing the global safety performance, and minimizing the global load deviation as the objective functions, and taking the requirement that high safety risk work orders must be matched with the maintenance persons in charge in the top 20% of safety performance as the constraint condition, the optimal solution set is output. The work order matching module sorts the maintenance persons in charge in the optimal solution set according to their ability priorities and provides them to the user.

[0013] In some embodiments, for the case of multiple work orders, the work order matching module sorts the work orders according to their urgency. According to the sorting from high to low urgency of the work orders, the work order matching module screens out the maintenance persons in charge who meet the time window and skill requirements of each work order, forming multiple sets of matching schemes for work orders and maintenance persons in charge. The work order matching module uses a multi-objective optimization algorithm to calculate the matching degree of each set of matching schemes. Taking maximizing the global skill matching degree, maximizing the global safety performance, and minimizing the global load deviation as the objective functions, and taking the requirement that high safety risk operation work orders must be matched with the maintenance persons in charge in the top 20% of safety performance as the constraint condition, the optimal solution set of the matching schemes is output. The work order matching module checks whether the time of the maintenance persons in charge in the matching schemes in the optimal solution set conflicts and whether the number of work orders exceeds the maximum carrying capacity. It corrects or deletes the schemes with time conflicts or exceeding the maximum carrying capacity, calculates the priority of the modified optimal solution set of matching schemes using the TOPSIS method of the ability matrix construction module, and provides them to the user according to the priority sorting.

[0014] In some embodiments, the objective function for maximizing the global skill matching degree of the work order matching module is as shown in formula (9): where k is the number of work orders, S(Pi) is the skill score of the maintenance person in charge Pi, and the calculation method is shown in formula (1); C(Ti) is the complexity of work order Ti, and the calculation method is shown in formula (7); E(Pi) is the experience value of the maintenance person in charge Pi, and the value range is (0, 1); w1 is the dynamic weight of the entropy weight method for the skill score of the maintenance person in charge, and w2 is the dynamic weight of the entropy weight method for the experience value of the maintenance person in charge, and the calculation method is shown in formula (6).

[0015] The objective function for maximizing the global safety performance is as shown in formula (10): Among them, k is the number of work orders; SAF(Pi) is the unitization of safety performance, which is the value calculated by formula (4) divided by 100, and its value range is (0, 1). The higher the safety performance value, the safer it is; L(Pi) is the load factor of the maintenance person in charge Pi, which is calculated by formula (3).

[0016] The objective function for minimizing the global load deviation is as shown in formula (11): Among them, m is the number of maintenance persons in charge, Nj is the number of work orders assigned to the maintenance person in charge Pj, and Qj is the maximum task capacity of the maintenance person in charge Pj.

[0017] In some embodiments, the system is also provided with a dynamic feedback module. The dynamic feedback module collects work order execution data in real time, compares the work order completion time and the planned time, the number of work order rework times and safety event records, and feeds back the abnormal conditions of the work order; the dynamic feedback module calculates the entropy value and weight of the four-dimensional ability index by calling the ability matrix construction module according to the historical task execution results, and dynamically adjusts the weight of the four-dimensional ability index; the dynamic feedback module reads the current load factor of the maintenance person in charge and displays it to the user in the form of a heat map. For the maintenance person in charge with a load factor greater than 85%, the dynamic feedback module gives an early warning; the dynamic feedback module compares the skill combination required by the work order and the skill qualifications of the maintenance person in charge, analyzes the ability short board of the maintenance person in charge, and automatically generates a report.

[0018] The present invention proposes a matching method for work leaders based on a personnel ability matrix. The method includes: Step 1: The work order feature extraction module analyzes the feature parameters of the new work order, reads the information of the work order, and establishes a work order feature model according to three indicators of complexity, urgency, and safety risk; Step 2: The work order matching module sorts the work orders according to the urgency of the work orders, and screens the maintenance persons in charge who meet the work order time window and skill requirements; Step 3: The ability matrix construction module quantifies the four-dimensional ability index of the maintenance persons in charge according to the skill requirements of the work order and constructs a decision matrix; Step 4: The work order matching module calculates the matching degree of each set of solutions according to the decision matrix and the work order feature model, outputs the solution set of the optimal matching solutions. The work order matching module checks whether the time of the maintenance persons in charge in the matching solutions in the optimal solution set conflicts and whether the number of work orders exceeds the maximum capacity. Modify or delete the matching solutions with time conflicts or exceeding the maximum capacity, calculate the priority of the modified matching solutions by using the TOPSIS method of the ability matrix construction module, and provide them to the user according to the priority ranking; Step 5: The dynamic feedback module reads the current load factor of the maintenance person in charge and displays it to the user in the form of a heat map. For maintenance persons in charge with a load factor greater than 85%, the dynamic feedback module gives an early warning.

[0019] Implementing the present invention has the following beneficial effects: The present invention provides a work person in charge matching system and matching method based on a personnel ability matrix. The system constructs a decision matrix through an ability matrix construction module, and a work order matching module matches work orders and persons in charge, achieving an improvement in work order matching efficiency. The time-consuming for work order allocation is shortened from an average of 2 hours manually to within 5 minutes, while also improving the matching accuracy rate. The system evaluates the safety risks of work orders through a work order feature extraction module, and high-risk work orders are automatically assigned to highly compliant persons in charge, effectively reducing the accident rate. The system monitors the load factor of maintenance persons in charge, avoids task overload, and improves resource utilization. Description of the Drawings

[0020] Figure 1 It is a flowchart of a work person in charge matching system based on a personnel ability matrix proposed in an embodiment of the present invention; Figure 2 It is a visual early warning flowchart of the dynamic feedback module of a work person in charge matching system based on a personnel ability matrix proposed in an embodiment of the present invention; Figure 3 It is a flowchart of ability matrix modeling of a work person in charge matching system based on a personnel ability matrix proposed in an embodiment of the present invention; Detailed Embodiments

[0021] The technical solutions of the present invention will be clearly and completely described below in conjunction with the drawings and detailed embodiments. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention fall within the scope of protection of the present invention.

[0022] As Figures 1 to 3 shown, the present invention provides a work person in charge matching system based on a personnel ability matrix, which includes a work order database, a personnel ability file library, a historical task library, an ability matrix construction module, a work order feature extraction module, a work order matching module, and a dynamic feedback module.

[0023] The work order database stores the historical data of nuclear power maintenance work orders. By performing natural language processing (NLP) on the unstructured text of nuclear power maintenance work orders (such as fault descriptions), the work order database extracts keywords and maps them to standardized tags (for example, "pump body leakage" → mechanical failure → complexity +2). Its standardized tags include fields such as work order ID, equipment type (such as reactor coolant pump, steam generator), fault type (instrument control fault, mechanical fault, electrical fault), work order complexity (low / medium / high), urgency (level 1-10), safety risk level (radiation level, operation specification level), required skill tags (such as "technician", "senior worker"), etc. The personnel ability profile database records the static data (qualification certificates, personal maximum task capacity) and dynamic data (current task volume, historical task completion rate, violation records) of the maintenance responsible person. The historical task database stores the execution result data of each work order, including completion time, number of reworks, and safety accident records.

[0024] The ability matrix construction module quantifies the four-dimensional ability indicators of the maintenance responsible person by reading the information in the database and constructs a decision matrix, and uses the TOPSIS method to dynamically assign weights to calculate the ability priority of each responsible person. Specifically, it includes: Static dimension: The index quantification of skill level. The ability matrix construction module calculates the skill scores according to the qualification certificate level (primary / intermediate / advanced) in the static data of the personnel ability profile database, as shown in formula (1): (1) Among them, the weight of the certificate level is specifically: primary = 0.3, intermediate = 0.6, advanced = 1.0; the weight of the skill type is specifically: core skill = 0.7, auxiliary skill = 0.3.

[0025] Dynamic dimension: The index quantification of experience value. The ability matrix construction module calculates the success rate by classifying and counting according to the equipment type and fault type based on the execution result data of each work order in the historical task database, and obtains the experience value, as shown in formula (2): Experience value (for a certain type of equipment) = (Number of successful equipment maintenance work orders / Total number of equipment maintenance work orders) (2) Real-time dimension: The index quantification of dynamic load. The ability matrix construction module collects the current task volume (such as the number of ongoing work orders) and the estimated available working hours (based on the schedule) of the maintenance responsible person in real time every 30 minutes, calculates the load coefficient, and calculates the available working hours within the next N hours. The load coefficient calculation formula is as shown in formula (3): Constraint Dimension: Quantification of indicators for safety performance. The ability matrix construction module reads violation records (such as not wearing protective clothing, operation overtime) and safety accident level deductions in the personnel ability database, calculates the safety performance, and the safety performance is specifically as shown in formula (4). Among them, the grade coefficient is specifically: minor = 1, general = 3, serious = 10.

[0026] After obtaining the four-dimensional ability indicators of the above maintenance person in charge, the ability matrix construction module constructs a decision matrix. Each row of this matrix represents a person in charge, and each column represents the four-dimensional ability indicator values. Then the ability matrix construction module calculates the entropy value of the ability indicator and determines the ability indicator weight to avoid subjective human intervention, specifically as shown in formula (5) and formula (6): Among them, e j is the entropy value of the j-th ability indicator, and the value range is [0, 1]. The larger the e j value, the more dispersed the information of this indicator and the lower the importance; m is the number of persons in charge, and p ij is the proportion of the score of the i-th person in charge on the j-th ability indicator to the total score of all persons in charge on this indicator. w j is the calculated weight of the j-th ability indicator, and the value range is [0, 1]. The sum of the calculated weights of all ability indicators is 1; e j is the entropy value of the j-th ability indicator, and n is the number of ability indicators. After obtaining the ability indicator weight, the ability matrix construction module calculates the ability priority of each person in charge through the TOPSIS method. The ability priority is used as an evaluation criterion for the rationality of the work order matching plan. In the TOPSIS method, by standardizing the decision matrix and multiplying it by the ability indicator weight, a weighted standardized decision matrix is obtained. Then, by determining the ideal solution through the standardized decision matrix, the distance and relative closeness between the four-dimensional ability indicators of each maintenance person in charge and the ideal solution are finally calculated. The closer the relative closeness is to 1, the closer the person in charge is to the most ideal solution, so as to judge the ability priority of each maintenance person in charge.

[0027] The work order feature extraction module constructs a feature parameter analysis for the new work order, reads the information of the work order, evaluates the work order from three indicators, and establishes a work order feature model according to the three indicators, specifically including: The work order feature extraction module quantifies the work order complexity, and the complexity is calculated according to the number and level of required skill combinations, specifically as shown in formula (7): (7) Among them, the specific skill level requirements are: beginner = 0.3, intermediate = 0.6, advanced = 1.0; the specific weights of skill types are: core skills = 0.7, auxiliary skills = 0.3.

[0028] The urgency of the work order is determined according to the work classification, and the work classification is determined by the work characteristic parameters of the work order. The work levels are divided into levels 0, 1, 2, 3, and 4, corresponding to work order urgencies of 1, 0.8, 0.6, 0.4, and 0.2 respectively.

[0029] The work order feature extraction module quantifies the safety risk of the work order. The safety risk is assigned values according to the radiation level and the operation specification level, specifically as shown in formula (8): (8) The radiation level coefficient is determined according to the radiation level of the operation area, which is a characteristic parameter of the work order. According to the general standards of nuclear power plants, the radiation level coefficients are divided into the red zone, orange zone, yellow zone, green zone, and zero-dose zone. The radiation level coefficient of the red zone is 1, the radiation level coefficient of the orange zone is 0.75, the radiation level coefficient of the yellow zone is 0.5, the radiation level coefficient of the green zone is 0.25, and the radiation level coefficient of the zero-dose zone is 0; the operation compliance coefficient depends on the human factor prevention level of the work order, which is a characteristic parameter of the work order. The values of the operation compliance coefficient are: 1, 0.8, 0.6, 0.4, 0.2.

[0030] After the work order feature extraction module evaluates the work order in the above three dimensions, a work order feature model is constructed according to the quantified values of the three dimensions.

[0031] The work order matching module reads the work order information and work order feature model of the work order feature extraction module to form a work order set T = {T1, T2,..., Tk}. The work order set includes the complexity, urgency, safety risk, and time window of each work order. The time window is the time period required for each work order.

[0032] The work order matching module reads the decision matrix of the ability matrix construction module to form a personnel set P = {P1, P2,..., Pm}. The personnel set includes the four-dimensional ability indicators of each maintenance person in charge and the available time interval of the maintenance person in charge.

[0033] For a single work order, the work order matching module screens out the maintenance responsible persons who meet the time window and skill requirements of the work order according to the work order requirements. The multi-objective optimization (NSGA-II) algorithm is used to calculate the matching degree between the work order and the maintenance responsible persons who meet the conditions of the work order. The objective functions are to maximize the global skill matching degree, maximize the global safety performance, and minimize the global load deviation. The constraint condition is that high safety risk work orders must be matched with the maintenance responsible persons in the top 20% of safety performance. Among them, the work order risk belongs to the characteristic parameters of the work order, which are divided into: first-level high-risk operations, second-level high-risk operations, and operations without safety risks according to nuclear power work orders. The optimal solution set is output, and the work order matching module sorts the maintenance responsible persons in the optimal solution set according to the ability priority of the maintenance responsible persons and provides them to the user.

[0034] For multiple work orders, the work order matching module sorts the work order urgency. According to the sorting from high to low urgency, the work order matching module screens out the maintenance responsible persons who meet the time window and skill requirements of each work order to form multiple sets of matching solutions. The work order matching module uses the multi-objective optimization (NSGA-II) algorithm to calculate the matching degree of each set of solutions. The objective functions are to maximize the global skill matching degree, maximize the global safety performance, and minimize the global load deviation. The constraint condition is that high safety risk work orders must be matched with the maintenance responsible persons in the top 20% of safety performance. The optimal solution set of the matching solutions is output. The work order matching module checks whether the time of the maintenance responsible persons in the matching solutions in the optimal solution set conflicts and whether the number of work orders exceeds the maximum carrying capacity. It corrects or deletes the solutions with time conflicts or exceeding the maximum carrying capacity, and calculates the priority of the modified matching solutions using the TOPSIS method of the ability matrix construction module, and provides them to the user according to the priority sorting. The ability priority sorting of the maintenance responsible persons calculated according to the TOPSIS method is recommended to the user.

[0035] Among them, the objective function for maximizing the global skill matching degree is as shown in formula (9): Among them, k is the number of work orders, S(Pi) is the skill score of the maintenance responsible person Pi, and the calculation method is shown in formula (1); C(Ti) is the complexity of the work order Ti, and the calculation method is shown in formula (7); E(Pi) is the experience value of the maintenance responsible person Pi, and the value range is (0, 1); w1 is the dynamic weight of the entropy weight method for the skill score of the maintenance responsible person, and w2 is the dynamic weight of the entropy weight method for the experience value of the maintenance responsible person, and the calculation method is shown in formula (6).

[0036] The objective function for maximizing the global safety performance is as shown in formula (10): Among them, k is the number of work orders; SAF(Pi) is the unitization of safety performance, which is the value calculated by formula (4) divided by 100, and its value range is (0, 1). The higher the safety performance value, the safer it is; L(Pi) is the load factor of the maintenance person in charge Pi, and its value range is (0, 1), which is calculated by formula (3).

[0037] The objective function for minimizing the global load deviation is as shown in formula (11): Among them, m is the number of maintenance persons in charge, Nj is the number of work orders assigned to the maintenance person in charge Pj, and Qj is the maximum task capacity of the maintenance person in charge Pj.

[0038] The dynamic feedback module collects the work order execution data in real time, compares the work order completion time and the planned time, the number of work order rework times (reflecting lack of skills or experience) and safety event records (such as radiation over - standard warnings). For maintenance persons in charge who have timed out or reworked continuously twice, an early warning of ability matrix degradation is triggered. The dynamic feedback module calls the ability matrix construction module to calculate the entropy value and weight of the four - dimensional ability index based on the historical task execution results over a period of time, and dynamically adjusts the weight of the four - dimensional ability index. For example, due to multiple radiation accidents in a certain power station, the system automatically calculates the entropy value and weight and increases the weight of safety performance. The dynamic feedback module also reads the current load factor of the maintenance person in charge and displays it to the user in the form of a heat map. For maintenance persons in charge with a load factor greater than 85%, the dynamic feedback module gives an early warning. The dynamic feedback module compares the skill combination required by the work order and the skill qualifications of the maintenance person in charge, analyzes the ability short - board of the maintenance person in charge, and automatically generates a report. For example, 10% of the persons in charge lack advanced radiation protection skills, and it is recommended to conduct targeted training.

[0039] The present invention proposes a method for matching work leaders based on a personnel ability matrix, and the method includes: Step 1: The work order feature extraction module analyzes the feature parameters of the new work order, reads the information of the work order, and establishes a work order feature model according to three indicators of complexity, urgency, and safety risk; Step 2: The work order matching module sorts the work orders according to the urgency of the work orders, and screens the maintenance persons in charge who meet the work order time window and skill requirements; Step 3: The ability matrix construction module quantifies the four - dimensional ability index of the maintenance person in charge according to the skill requirements of the work order and constructs a decision matrix; Step 4: The work order matching module calculates the matching degree of each set of solutions according to the decision matrix and the work order feature model. The work order matching module checks whether there are time conflicts among the maintenance responsible persons in the matching solutions in the optimal solution set and whether the number of work orders exceeds the maximum load capacity. It corrects or deletes the matching solutions with time conflicts or exceeding the maximum load capacity, calculates the priority of the modified matching solutions using the TOPSIS method of the capability matrix construction module, and provides them to the user according to the priority ranking; Step 5: The dynamic feedback module reads the current load factor of the maintenance responsible persons and displays it to the user in the form of a heat map. For the maintenance responsible persons with a load factor greater than 85%, the dynamic feedback module gives an early warning.

[0040] The present invention proposes a matching system for work responsible persons based on a personnel capability matrix. The specific embodiments of the system are as follows: Embodiment 1: A sudden high-voltage electrical fault occurs in the reactor coolant pump of Unit 1 of a nuclear power plant, and emergency repair is required.

[0041] ‌‌The work order feature extraction module analyzes the work order and finds that the complexity = 0.92 (requiring high-voltage electrical senior + mechanical intermediate skills), the urgency = 9 / 10, and the safety risk = 8 / 10 (high-radiation area).

[0042] The work order matching module screens the responsible persons and pre-screens 5 qualified responsible persons, among which 3 have a safety performance ≥ 90 points.

[0043] The work order matching module reads the decision matrix constructed by the capability matrix construction module and calculates the matching between the work order and the maintenance responsible persons: Responsible person A: Skill matching degree = 0.95, load factor = 65%, safety performance = 95, TOPSIS capability priority 0.93. Responsible person B: Skill matching degree = 0.88, load factor = 45%, safety performance = 92, TOPSIS capability priority 0.89.

[0044] The work order matching module outputs the result and recommends responsible person A as the first choice, and prompts "Load is safe, and the success rate of historical similar work orders is 100%" according to the information of responsible person A.

[0045] Responsible person A completes the task within 4 hours without rework. The dynamic feedback module dynamically adjusts the weights of the four-dimensional capability indicators and increases his experience value from 92 to 94.

[0046] Embodiment 2: The nuclear power plant needs to process 3 work orders simultaneously as follows: T1 (high urgency): Time window [9:00 - 12:00], complexity = 4, safety risk = 8.

[0047] T2 (Medium Urgency): Time window [10:00 - 15:00], Complexity = 3, Safety Risk = 5.

[0048] T3 (Low Urgency): Time window [14:00 - 18:00], Complexity = 2, Safety Risk = 3.

[0049] The candidate maintenance responsible persons are as shown in Table 1 below: Table 1 Maintenance Responsible Persons Table The work order matching module sorts the work orders according to the urgency. The work order priorities are sorted as: T1 > T2 > T3. The work order matching module uses a multi - objective optimization algorithm to generate the following optimal solution set: Solution 1: T1 → A, T2 → B, T3 → C (Load: A = 1, B = 1, C = 1; No time conflict).

[0050] Solution 2: T1 → A, T2 → C, T3 → A (Load: A = 2, C = 1; A has time overlap in the time windows of T1 and T3).

[0051] The work order matching module discovers the time conflict of A in Solution 2, so it re - assigns work order T3 to C.

[0052] Calculated by the TOPSIS method, the priority of Solution 1 is higher than that of the corrected Solution 2. The work order matching module finally recommends Solution 1.

[0053] The above embodiments only represent several implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several deformations and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention should be subject to the appended claims.

Claims

1. A work leader matching system based on a personnel ability matrix, characterized in that, The system includes a database, a competency matrix construction module, a work order feature extraction module, and a work order matching module. The database includes work order historical data, maintenance person-in-charge data, and historical task data. The competency matrix construction module quantifies the four-dimensional competency indicators of the maintenance person-in-charge by reading the information in the database and constructs a decision matrix. The work order feature extraction module analyzes the feature parameters of the work order and establishes a work order feature model. The work order matching module reads the information in the database, the decision matrix of the competency matrix construction module, and the work order feature model of the work order feature extraction module to match the maintenance person-in-charge with the work order.

2. The work leader matching system based on a personnel ability matrix according to claim 1, wherein The database includes a work order database, a personnel competency archive database, and a historical task database. The work order database stores the historical data of nuclear power maintenance work orders. The work order database performs natural language processing on the unstructured text of nuclear power maintenance work orders, extracts keywords, and maps them to standardized tags. The personnel competency archive database records the qualification certificates, personal maximum task capacity, current task volume, historical task completion rate, and violation records of the maintenance person-in-charge. The historical task database stores the execution result data of each work order.

3. The work leader matching system based on a personnel ability matrix according to claim 2, characterized in that, The specific steps for the competency matrix construction module to quantify the four-dimensional competency indicators of the maintenance person-in-charge and construct a decision matrix are as follows: The competency matrix construction module calculates the skill scores according to the qualification certificate levels in the personnel competency archive database, as shown in formula (1): Among them, the specific weights of the certificate levels are: the primary certificate level is 0.3, the intermediate certificate level is 0.6, and the advanced certificate level is 1.0; the specific weights of the skill types are: the core skill is 0.7, and the auxiliary skill is 0.

3. Based on the execution result data of each work order in the historical task database, the competency matrix construction module classifies and statistically calculates the success rate of the maintenance person-in-charge according to the equipment type and fault type, and calculates the experience value, as shown in formula (2): The competency matrix construction module collects the current task volume and the estimated available working hours of the maintenance person-in-charge, calculates the load factor, and the calculation formula of the load factor is shown in formula (3): The competency matrix construction module reads the violation records in the personnel competency archive database and calculates the safety performance. The safety performance is shown in formula (4): Among them, the specific level coefficients are: a minor violation is 1, a general violation is 3, and a serious violation is 10. After obtaining the four-dimensional competency indicators of the maintenance person-in-charge, the competency matrix construction module constructs a decision matrix. Each row of the decision matrix represents a maintenance person-in-charge, and each column represents the four-dimensional competency indicator values.

4. The work leader matching system based on a personnel ability matrix according to claim 3, characterized in that The competency matrix construction module calculates the entropy value of the competency indicators according to the decision matrix and determines the weights of the competency indicators. Then, according to the weights of the competency indicators and the TOPSIS method, it calculates the competency priority of each person-in-charge. The calculation formula of the entropy value of the competency indicators is specifically shown in formula (5): Among them, e j is the entropy value of the j-th ability index; m is the number of responsible persons, p ij is the proportion of the value of the i-th responsible person on the j-th ability index to the total value of all responsible persons of the index; The calculation formula of the weights of the competency indicators is specifically shown in formula (6): Among them, w j is the calculation weight of the j-th ability index; e j is the entropy value of the j-th ability index, and n is the number of ability indices.

5. The job leader matching system based on the personnel ability matrix according to claim 4, wherein The work order feature extraction module reads the information of the work order and establishes a work order feature model from three aspects: complexity, urgency, and safety risk. Among them, the calculation of the complexity by the work order feature extraction module is specifically shown in formula (7): （7） Among them, the specific skill level requirements are as follows: the primary skill level requirement is 0.3, the intermediate skill level requirement is 0.6, and the advanced skill level requirement is 1.0; the specific weights of skill types are as follows: core skills = 0.7, auxiliary skills = 0.3; The work order feature extraction module determines according to the work classification, and the work classification is determined by the work characteristic parameters of the work order. The work levels are divided into levels 0, 1, 2, 3, and 4. Among them, the work order urgency corresponding to level 0 work level is 1, the work order urgency corresponding to level 1 work level is 0.8, the work order urgency corresponding to level 2 work level is 0.6, the work order urgency corresponding to level 3 work level is 0.4, and the work order urgency corresponding to level 4 work level is 0.2; The work order feature extraction module quantifies the safety risk of the work order, and the safety risk is assigned according to the radiation level and the operation specification level, specifically as shown in formula (8): （8） Among them, the radiation level coefficient is determined according to the radiation level of the operation area and obtained from the characteristic parameters of the work order. According to the general standards of nuclear power plants, the radiation level coefficients are divided into the red zone, orange zone, yellow zone, green zone, and zero-dose zone. The radiation level coefficient of the red zone is 1, the radiation level coefficient of the orange zone is 0.75, the radiation level coefficient of the yellow zone is 0.5, the radiation level coefficient of the green zone is 0.25, and the radiation level coefficient of the zero-dose zone is 0; the operation standardization coefficient depends on the human factor prevention level of the work order and is obtained from the characteristic parameters of the work order.

6. The work leader matching system based on the personnel ability matrix according to claim 5, characterized in that For a single work order, the work order matching module screens out the maintenance responsible persons who meet the time window and skill requirements of the work order according to the work order requirements. Using a multi-objective optimization algorithm, the matching degree between each work order and the maintenance responsible persons meeting the work order conditions is calculated. Taking maximizing the global skill matching degree, maximizing the global safety performance, and minimizing the global load deviation as the objective functions, and taking the requirement that high safety risk work orders must be matched with the top 20% of the maintenance responsible persons with safety performance as the constraint condition, the optimal solution set is output. The work order matching module sorts the maintenance responsible persons in the optimal solution set according to their ability priorities and provides them to the user.

7. The work leader matching system based on a personnel ability matrix according to claim 6, wherein, For the case of multiple work orders, the work order matching module sorts the work order urgencies. According to the sorting of work order urgencies from high to low, the work order matching module screens out the maintenance responsible persons who meet the time window and skill requirements of each work order, forming multiple sets of matching schemes for work orders and maintenance responsible persons. The work order matching module uses a multi-objective optimization algorithm to calculate the matching degree of each set of matching schemes. Taking maximizing the global skill matching degree, maximizing the global safety performance, and minimizing the global load deviation as the objective functions, and taking the requirement that high safety risk work orders must be matched with the top 20% of the maintenance responsible persons with safety performance as the constraint condition, the optimal solution set of the matching schemes is output. The work order matching module checks whether the times of the maintenance responsible persons in the matching schemes in the optimal solution set conflict and whether the number of work orders exceeds the maximum carrying capacity, modifies or deletes the schemes with time conflicts or exceeding the maximum carrying capacity, calculates the priorities of the modified optimal solution set of matching schemes using the TOPSIS method of the ability matrix construction module, and provides them to the user according to the priority sorting.

8. The work leader matching system based on a personnel capability matrix according to claim 7, characterized in that, The objective function for maximizing the global skill matching degree of the work order matching module is as shown in formula (9): where k is the number of work orders, S(Pi) is the skill score of the maintenance person in charge Pi, and the calculation method is shown in formula (1); C(Ti) is the complexity of work order Ti, and the calculation method is shown in formula (7); E(Pi) is the experience value of the maintenance person in charge Pi, and the value range is (0, 1); w1 is the dynamic weight of the entropy weight method for the skill score of the maintenance person in charge, and w2 is the dynamic weight of the entropy weight method for the experience value of the maintenance person in charge, and the calculation method is shown in formula (6); The objective function for maximizing the global safety performance is as shown in formula (10): where k is the number of work orders; SAF(Pi) is the unitization of safety performance, which is the value obtained by formula (4) divided by 100, and the value range is (0, 1). The higher the safety performance value, the safer it is; L(Pi) is the load factor of the maintenance person in charge Pi, which is calculated by formula (3); The objective function for minimizing the global load deviation is as shown in formula (11): where m is the number of maintenance persons in charge, Nj is the number of work orders assigned to the maintenance person in charge Pj, and Qj is the maximum task carrying capacity of the maintenance person in charge Pj.

9. The work leader matching system based on the personnel ability matrix according to claim 8, wherein, The system also has a dynamic feedback module, which collects work order execution data in real time, compares the work order completion time and the planned time, the number of work order reworks and safety event records, and feeds back the abnormal conditions of the work order; The dynamic feedback module calculates the entropy value and weight of the four-dimensional ability index by calling the ability matrix construction module according to the historical task execution results, and dynamically adjusts the weight of the four-dimensional ability index; The dynamic feedback module reads the current load factor of the maintenance person in charge and displays it to the user in the form of a heat map. For the maintenance person in charge with a load factor greater than 85%, the dynamic feedback module gives an early warning; The dynamic feedback module compares the skill combination required by the work order and the skill qualifications of the maintenance person in charge, analyzes the ability short board of the maintenance person in charge, and automatically generates a report.

10. A method for matching a person in charge of work based on a personnel ability matrix, wherein the matching method uses the work leader matching system described in claim 1, characterized in that, The method includes: Step 1: The work order feature extraction module analyzes the feature parameters of the new work order, reads the information of the work order, and establishes a work order feature model according to the three indicators of complexity, urgency, and safety risk; Step 2: The work order matching module sorts the work orders according to the urgency of the work order, and screens the maintenance persons in charge who meet the time window and skill requirements of the work order; Step 3: The ability matrix construction module quantifies the four-dimensional ability index of the maintenance person in charge according to the skill requirements of the work order and constructs a decision matrix; Step 4: The work order matching module calculates the matching degree of each set of solutions according to the decision matrix and the work order feature model, outputs the solution set of the optimal matching solution, and the work order matching module checks whether the time of the maintenance person in charge in the matching solution in the optimal solution set conflicts and whether the number of work orders exceeds the maximum carrying capacity, modifies or deletes the matching solution with time conflict or exceeding the maximum carrying capacity, calculates the priority of the modified matching solution by using the TOPSIS method of the ability matrix construction module, and provides it to the user according to the priority ranking; Step 5: The dynamic feedback module reads the current load factor of the maintenance person in charge and displays it to the user in the form of a heat map. For the maintenance person in charge with a load factor greater than 85%, the dynamic feedback module issues a warning.

Images