Post competency evaluation system based on virtual reality technology

Through virtual reality technology, capturing and analyzing employee behavior data, evaluating decision-making and learning styles, and quantifying team interactions, the problems of bias and inaccuracy of evaluation results in traditional systems are solved, and more accurate job competency evaluation and personalized training are achieved. , improves team collaboration efficiency.

CN119026971APending Publication Date: 2024-11-26GUANGZHOU JINYUAN TECH DEV CO LTD
View PDF 0 Cites 12 Cited by

Patent Information

Application Number
CN202410891188.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-04
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

The traditional job competency evaluation system relies on quantitative and qualitative tools, and there are deviations and inaccuracies in the evaluation results, which makes it difficult to fully reflect the performance of employees in different situations, especially in the measurement of behavioral characteristics and psychological quality, which cannot accurately reflect the employees' Actual work ability and potential, and lack of comprehensive analysis of teamwork and interaction.

Method used

Adopt a job competency evaluation system based on virtual reality technology, obtain employee behavior data through the behavior data capture module, screen key time points for the behavior dynamic analysis module, evaluate decision-making quality of the decision, learn habit recognition module recognizes learning style, and interactive mode mapping Module analysis team interaction, employee capability difference analysis module quantitative collaboration and leadership ability, training effect verification module evaluation team collaboration.

Benefits of technology

It improves the accuracy and real-time performance evaluation, enhances the understanding of employee behavior dynamics, accurately evaluates decision-making ability, recognizes learning styles, promotes personalized training, improves team collaboration efficiency, and enhances the relevance and effectiveness of training.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119026971A_ABST
    Figure CN119026971A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of employee performance evaluation, in particular to a post competency evaluation system based on a virtual reality technology. The system comprises a behavior data capture module, a behavior dynamic analysis module, a decision ability evaluation module, a learning habit identification module, an interaction mode mapping module, an employee ability difference analysis module and a training effect verification module. According to the invention, the employee behavior data is obtained in the virtual reality, the innovation scheme improves the precision and real-time performance evaluation, the integrity and timeliness of the data are verified in real time by recording the position movement, the operation frequency and the decision duration of the employees, the core characteristics of the employee behaviors are further informed, and the performance evaluation accuracy is improved. According to the method, the dynamic understanding of employee behaviors is enhanced, a management layer is allowed to accurately analyze key time points influencing the employee behaviors, so that the behavior trend and the abnormal mode are identified, the decision-making ability of the employees is more accurately evaluated through behavior mode identification, and the decision-making quality is quantified by comparing the decision-making efficiency and the result.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of employee performance evaluation, and in particular to a job competency evaluation system based on virtual reality technology. Background Art

[0002] The field of employee performance evaluation technology focuses on using various tools and methods to systematically measure, analyze and optimize employee performance at work. It involves quantitative and qualitative evaluation techniques, and aims to provide continuous feedback and support to help employees improve their work efficiency and quality. Its main applications include goal setting, performance monitoring, capacity development, feedback and coaching, which helps determine whether employees have met the predetermined work standards and provides data support for human resource management, so as to make more scientific personnel decisions, such as promotion, training needs analysis and the formulation of employee incentive plans.

[0003] Among them, the job competency evaluation system is a system designed specifically to assess whether employees have the skills and knowledge required for their positions. The system measures employees' performance in specific professional skills, knowledge levels, behavioral characteristics, and psychological qualities by setting specific evaluation standards and tests. This evaluation is used to screen suitable candidates during the recruitment process, employee promotion assessment, and determine training and development needs. Effective job competency evaluation can help companies ensure that employees are highly matched with their positions, thereby improving work efficiency and team performance, while reducing talent loss.

[0004] Traditional evaluation systems mainly rely on quantitative and qualitative tools, such as questionnaires, interviews, and performance appraisal forms. They rely on subjective judgments and are prone to bias and inaccuracy in evaluation results. The existing technology lacks real-time and dynamic evaluation of employee behavior and decision-making ability, and it is difficult to fully reflect the performance of employees in different situations. Although the job competency evaluation system can evaluate employees' professional skills and knowledge levels, it has limitations in measuring behavioral characteristics and psychological qualities, and cannot accurately reflect employees' actual work ability and potential. The existing technology is relatively weak in the evaluation of teamwork and interaction, and lacks a comprehensive analysis of the role and influence of employees in a team environment, making it difficult to formulate effective team management strategies. It affects scientific decision-making in human resource management, such as promotion, training needs analysis, and the formulation of employee incentive plans, thereby affecting the overall performance and competitiveness of the enterprise. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a job competency evaluation system based on virtual reality technology.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: A job competency evaluation system based on virtual reality technology includes: The behavior data capture module records employee behavior data including location movement, operation frequency, and decision-making duration based on the employee behavior data obtained by the VR device, verifies the integrity and timeliness of the data, and obtains a set of behavior characteristics; The behavior dynamic analysis module screens the key time points that affect employee behavior based on the behavior feature set, calculates the behavior variables at the time points, analyzes the behavior trends and abnormal patterns, and obtains the behavior pattern indicators; The decision-making ability evaluation module conducts efficiency and result comparison analysis on the extracted decision points based on the behavioral pattern indicators, evaluates the employee's response speed and selection quality, and obtains a decision-making quality index; The learning habit identification module analyzes the decision frequency and complexity of employees based on the decision quality index, identifies the decision pattern, compares the deviation from the standard pattern, and obtains the learning style identification result; The interaction mode mapping module extracts the interaction data of individuals and groups based on the learning style identification results, draws a frequency graph of member interactions, and obtains the team interaction effectiveness; The employee capability difference analysis module records the interaction frequency and influence of employees based on the team interaction effectiveness, quantifies the collaboration and leadership capabilities of employees, simulates the work scenarios of target responsibilities, evaluates the professional skills and strategy application of employees in various positions, and generates customized training programs for each position; The training effect verification module sets team tasks in virtual reality based on the team interaction effectiveness, monitors the completion of tasks, evaluates the impact of teamwork on individual and collective performance, and obtains training application evaluation results.

[0007] As a further solution of the present invention, it is characterized in that: the behavioral feature set includes employee location coordinates, operation timestamps and decision reaction time, the behavioral pattern indicators include time-sensitive behavior points, speed changes, and path deviation indicators, the decision quality index includes decision speed index, decision accuracy index and decision consistency score, the learning style identification results include decision frequency measurement, problem solving complexity rating and standard model deviation comparison data, the team interaction effectiveness includes team member interaction frequency graph, role adaptability analysis results and teamwork efficiency evaluation results, the job customized training program includes management skills optimization, technical problem solution, and customer interaction efficiency optimization, and the training application evaluation results include task completion time records, team task completion, and individual performance data.

[0008] As a further solution of the present invention, the behavior data capturing module includes: The behavior monitoring submodule monitors the dynamic position of employees in the virtual environment in real time based on the employee behavior data obtained by the VR device, records the real-time reaction speed of employee operations, captures the key time points of decision-making events, and generates an initialization behavior data set; The data synchronization submodule performs standardization processing on the data based on the initialization behavior data set, converts the data format uniformly, eliminates errors and redundancies in the data, and generates formatted behavior data; The data verification submodule detects the integrity and timeliness of the data based on the formatted behavior data, updates the timestamp of the behavior information in real time, serializes and marks the data blocks, and generates a behavior feature set.

[0009] As a further solution of the present invention, the behavior dynamic analysis module includes: The behavior time point screening submodule screens the key time points that affect the employee's behavior based on the behavior feature set, matches the timestamp data with the behavior event, and analyzes the key time period according to the employee's position change and operation frequency in the VR environment to obtain a key time point set; The behavior variable calculation submodule calculates the behavior variables at the time points based on the key time point set, calculates the speed changes between various time points through time series, calculates the path deviation range through trajectory deviation, compares the real-time path with the preset path, and generates a behavior variable set; The behavior pattern analysis submodule performs cluster analysis of the behavior patterns based on the set of behavior variables, groups the behavior data, extracts the characteristic vectors of the behavior variables, analyzes the fit and differences between the variables, identifies the pattern associations between the behavior variables, detects abnormal behavior characteristics, and obtains behavior pattern indicators.

[0010] As a further solution of the present invention, the decision-making ability assessment module includes: The decision point extraction submodule analyzes the decision behavior data based on the behavior pattern indicators, marks the operation position and timestamp of the decision points, screens the key decision nodes that affect the overall decision, and obtains a decision point set; The efficiency comparison and analysis submodule performs efficiency analysis on the decision points based on the decision point set, calculates the reaction time of the decision points, evaluates the processing speed of the decision points, compares the reaction time of multiple decision points, and obtains decision efficiency data; The quality assessment submodule uses decision tree analysis technology based on the decision efficiency data to evaluate employees' response speed and selection quality, analyze the accuracy and consistency of decisions, compare the selection results of multiple decision points, evaluate employees' performance in the decision-making process, and obtain a decision quality index.

[0011] The decision tree analysis technique follows the formula: ; in, For collection and The entropy of is the target dataset, is the set of decision points, For decision point The following subset, For the subset The probability of the i-th element in , is the balancing factor, It is a decision point The quality rating of and Represents the number of elements in the dataset and subset respectively, is the number of decision points, is the number of categories, is the number of rating dimensions.

[0012] As a further solution of the present invention, the learning habit recognition module includes: The decision frequency analysis submodule records the time intervals of employee decisions based on the decision quality index, marks the time and frequency of decisions, and generates decision frequency data; The decision complexity assessment submodule analyzes and quantifies the number of variables and the degree of interaction at key decision points based on the decision frequency data, records the decision path, tracks the execution time and action behaviors required for employees to make decisions, and generates decision complexity data; The learning behavior classification submodule identifies the employee's decision-making pattern based on the decision complexity data, compares the deviation from the standard pattern, analyzes the difference in decision-making behavior and selection path, classifies the employee's learning behavior, and obtains the learning style identification result.

[0013] As a further solution of the present invention, the interactive mode mapping module includes: The interaction record submodule extracts individual and group interaction data based on the learning style identification results, collects employee interaction records in the virtual environment, records the number of members' interactions and interaction objects, and generates an interaction data set; The interaction frequency drawing submodule counts the interaction frequency of members based on the interaction data set, analyzes the time distribution of interaction events, records the interaction status in key time periods, and generates an interaction frequency graph; The team role analysis submodule analyzes the interaction relationship of employees in the virtual environment based on the interaction frequency graph, assigns team roles, and generates team interaction effectiveness by evaluating the number and breadth of interactions, analyzing the influence and role distribution among members.

[0014] As a further solution of the present invention, the employee capability difference analysis module includes: The employee classification submodule classifies employees into management personnel, technical personnel and grassroots employees according to the performance data based on the team interaction effectiveness and the performance of the employees in the simulation environment, and uses statistical analysis to set classification thresholds to generate an employee category list; The skill assessment submodule simulates target responsibility scenarios of multiple types of employees based on the employee category list, verifies the professional skills of multiple types of employees through real-time operations, including problem solving, decision making and customer service capabilities, and generates skill analysis results; The strategy formulation submodule formulates exclusive training strategies based on the skill analysis results and the employee's classification results and professional skills, including setting training goals, selecting training methods, arranging training schedules, adjusting training content through a training effect feedback loop, and generating job-customized training plans.

[0015] As a further solution of the present invention, the training effect verification module includes: The team collaboration task submodule customizes the training program based on the positions, simulates the team collaboration tasks in virtual reality, assigns roles to team members, defines the responsibilities and tasks of the roles, establishes team collaboration task scenarios, and generates team task data; The real-time monitoring submodule monitors the collaboration and task completion of employees in real-time operations based on the team task data, records the task progress and interactive behaviors of members, records the number of abnormal operations and errors of employees, and analyzes the number of collaborations and interaction frequencies to generate real-time operation records; The collaboration impact analysis submodule uses a random forest algorithm based on the real-time operation records to analyze the impact of team collaboration on individual learning and overall performance, evaluate the performance of members in the task, compare the task completion of individuals and teams, and obtain the training application evaluation results by analyzing the optimization of learning effects and changes in the overall team effectiveness caused by collaboration.

[0016] The random forest algorithm follows the formula: ; in, To predict the results, is the number of decision trees, For the The prediction function of a decision tree, is the input feature, For the The set of random parameters associated with each decision tree, For the The weight of a decision tree, is the control parameter of the decision tree depth, is the minimum number of samples for splitting a decision tree node, is the balance coefficient, is the regularization parameter.

[0017] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by acquiring employee behavior data in virtual reality, the innovative solution improves the accuracy and real-time performance evaluation. By recording the employee's position movement, operation frequency and decision-making time, the integrity and timeliness of the data are verified in real time, and further insight into the core characteristics of employee behavior is obtained. It not only enhances the understanding of employee behavior dynamics, but also allows management to accurately analyze the key time points that affect employee behavior, thereby identifying behavioral trends and abnormal patterns. Behavioral pattern recognition helps to more accurately assess the decision-making ability of employees and quantify the quality of decisions by comparing decision-making efficiency and results. By analyzing the frequency and complexity of decisions, different learning styles are identified, further facilitating the formulation of personalized training and development plans. The personalized approach not only increases the relevance and effectiveness of training, but also improves the effectiveness of team collaboration by mapping and analyzing interaction patterns within the team. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a system flow chart of the present invention; Figure 2 It is a schematic diagram of the system framework of the present invention; Figure 3 is a flow chart of the behavior data capture module of the present invention; Figure 4 It is a flow chart of the behavior dynamic analysis module of the present invention; Figure 5 It is a flow chart of the decision-making ability assessment module of the present invention; Figure 6 is a flow chart of the learning habit recognition module of the present invention; Figure 7 It is a flow chart of the interactive mode mapping module of the present invention; Figure 8 It is a flow chart of the employee capability difference analysis module of the present invention; Fig. 9 This is a flow chart of the training effect verification module of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0020] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0021] Embodiment 1 See also Figure 1 to Figure 2 , a job competency evaluation system based on virtual reality technology includes: The behavior data capture module obtains employee behavior data based on VR devices, records data including location movement, operation frequency, and decision duration, synchronizes the data to the central database, verifies the integrity and timeliness of the data, and obtains a set of behavior characteristics; The behavior dynamic analysis module screens the key time points that affect employee behavior based on the behavioral feature set, calculates the behavioral variables at the time points, including speed changes and path deviations, performs cluster analysis of behavioral patterns, identifies behavioral trends and abnormal patterns, and obtains behavioral pattern indicators; The decision-making ability assessment module extracts key decision points in decision-making based on behavioral pattern indicators, conducts comparative analysis of the efficiency and results of the decision points, evaluates the employee's response speed and selection quality, and obtains the decision quality index; The learning habit identification module analyzes the decision frequency and complexity of employees based on the decision quality index, identifies decision patterns, compares deviations from standard patterns, and classifies employees’ learning behaviors to obtain learning style identification results; The interactive mode mapping module extracts individual and group interaction data based on the learning style identification results, draws a frequency graph of member interactions, analyzes team role allocation and influence based on the interactive relationships of employees in the virtual environment, and obtains team interaction effectiveness; The employee capability difference analysis module is based on team interaction effectiveness. It analyzes the interaction data of team members in a virtual environment, records the interaction frequency and influence of employees, and quantifies the collaboration and leadership capabilities of employees. It simulates the work scenarios of target responsibilities, evaluates the professional skills and strategy application of employees in various positions, and generates customized training programs for each position. The training effectiveness verification module customizes training plans based on positions, simulates team collaboration tasks in virtual reality, monitors employees' collaboration and task completion in real-time operations, analyzes the impact of team collaboration on individual learning and overall performance, and obtains training application evaluation results.

[0022] The behavioral feature set includes employee location coordinates, operation timestamps and decision response time. The behavioral pattern indicators include time-sensitive behavior points, speed changes, and path deviation indicators. The decision quality index includes decision speed index, decision accuracy index and decision consistency score. The learning style identification results include decision frequency measurement, problem solving complexity rating and standard model deviation comparison data. The team interaction effectiveness includes team member interaction frequency graph, role adaptability analysis results and teamwork efficiency evaluation results. The job customized training program includes management skills optimization, technical problem solution method, and customer interaction efficiency optimization. The training application evaluation results include task completion time records, team task completion degree, and individual performance data.

[0023] Please refer to Figure 3 and Figure 2 , the behavioral data capture module includes: The behavior monitoring submodule monitors the dynamic position of employees in the virtual environment in real time based on the employee behavior data obtained by the VR device, records the real-time reaction speed of employee operations, captures the key time points of decision-making events, and generates the initialization behavior data set. The execution process is as follows; The behavior monitoring submodule records the movement trajectory of each employee in the virtual environment in detail based on the employee behavior data obtained by the VR device, uses time series analysis to calculate the continuity and smoothness of the action, compares the preset action response standards, and identifies and marks the abnormal points in the behavior data. The abnormal points include abnormal dwell time, sudden changes in speed, etc. The behavior data set is further screened to remove noise data, such as short-term irrelevant actions or data deviations caused by equipment errors. After screening and preliminary analysis, the data is encoded into a format suitable for further processing. Each data point is equipped with a timestamp and location mark to generate an initialized behavior data set. The formula is: ; in, represents the central tendency of decision-making time points, For the The time point of the next decision, is the time interval from the last decision, is the attenuation coefficient.

[0024] The data synchronization submodule standardizes the data based on the initialization behavior data set, converts the data format uniformly, eliminates errors and redundancies in the data, and generates the execution flow of formatted behavior data as follows; The data synchronization submodule audits the raw data based on the initialized behavioral data set, checks the consistency and integrity of the data, and identifies and corrects data errors, such as timestamp errors and deviations in location data. The submodule applies data cleaning techniques to remove non-task-related behavioral records and obvious errors, such as abnormal data caused by equipment failure. Each data point is verified to ensure that it reflects real employee behavior. The submodule normalizes the data and standardizes data records from different devices or under different conditions so that the data sets are comparable in subsequent analysis. The standardization process includes converting all location data into a unified coordinate system, adjusting the time data to a unified time zone, and generating formatted behavioral data. The formula is: ; in, represents a new data point, is the original data point, is the average value of the data points, is the smoothing coefficient.

[0025] The data verification submodule detects the integrity and timeliness of the data based on the formatted behavior data, updates the timestamp of the behavior information in real time, serializes and marks the data blocks, and generates the execution process of the behavior feature set as follows; The data validation submodule performs integrity checks on each data point based on formatted behavioral data to ensure that all required fields are filled and no data is missing. The module detects the timeliness of the data to ensure the accuracy and timeliness of the data timestamp. If a significant deviation is found between the timestamp and the system time, necessary corrections are made. The module serializes and marks the data, assigns a unique identifier to each data block, and records the generation time and last modification time of the data block, tracks historical changes in the data, ensures data version management, and generates a set of behavioral features. The formula is: ; in, represents the generated eigenvalue, For the The weight of the data point, is the probability of a data point appearing in its category, is the total number of data points.

[0026] See also Figure 4 and Figure 2 , the behavioral dynamics analysis module includes: The behavior time point screening submodule screens the key time points that affect employee behavior based on the behavior feature set, matches the timestamp data with the behavior events, and analyzes the key time periods based on the employee's position changes and operation frequency in the VR environment. The execution process of the key time point set is as follows; The behavior time point screening submodule is based on a set of behavioral features. The module uses a screening algorithm to identify time points associated with important behavioral events from a large amount of timestamp data. The correlation between each timestamp and the behavioral event is quantified, and specific indicators are used to determine the importance of each time point. The module further analyzes the time points based on the position changes and operation frequency of employees in the virtual environment to screen out significant time periods in the behavioral pattern. For position changes, the module analyzes the path and speed of employee movement to identify time points with sudden position changes or abnormal stays. For operation frequency, the module calculates the interval time and frequency of operations to identify abnormal or critical operations in the behavior pattern and obtain a set of key time points. The formula is: ; in, Indicates time point The influence of is the sensitivity parameter of the time point, is the central position parameter of the time point.

[0027] The behavior variable calculation submodule calculates the behavior variables at the time points based on the key time point set, calculates the speed changes between various time points through time series, and calculates the path deviation range through trajectory deviation, compares the real-time path with the preset path, and generates the behavior variable set. The execution process is as follows; The behavior variable calculation submodule is based on a set of key time points. It uses the data in the set of key time points to obtain the position change between adjacent time points through the difference method, and then calculates the speed. The submodule analyzes the deviation between the employee's real-time movement trajectory and the preset trajectory, and uses the Euclidean distance to quantify the degree of deviation at each time point. In order to comprehensively evaluate the employee's behavior variables, the module combines the speed and path deviation data and uses a weighted method to calculate the comprehensive indicators of the behavior variables. Accurately capture and quantify the behavioral characteristics of employees at key time points to generate a set of behavior variables. The formula is: ; in, Indicates The comprehensive index of behavioral variables at each time point, is the weight coefficient, and represent the changes in position and time respectively, , is the coordinate point of the employee’s actual trajectory, , It is the coordinate point of the preset trajectory.

[0028] The behavior pattern analysis submodule performs cluster analysis of behavior patterns based on the behavior variable set, groups the behavior data, extracts the characteristic vectors of the behavior variables, analyzes the fit and difference between the variables, identifies the pattern association between the behavior variables, and detects abnormal behavior characteristics. The execution process of obtaining the behavior pattern indicators is as follows; The behavior pattern analysis submodule is based on a set of behavior variables and uses multi-dimensional feature extraction technology to convert each behavior data point into a feature vector. The feature vector contains detailed information about employee behavior, such as movement speed, operation frequency, reaction time, etc. By applying clustering algorithms, such as K-means or hierarchical clustering, the submodule groups the behavior data according to the similarity of the feature vectors, revealing the commonalities and differences in many behaviors. After cluster analysis, the data within each group is fit analyzed, and the correlation and covariance of the variables within each group are calculated to evaluate the consistency of the behavior pattern and the effectiveness of the group division, and obtain the behavior pattern indicators. Improve the formula for pattern association analysis: ; in, Representation variables and The correlation coefficient between and The variables are and In the The value of the data point, and is a variable and The average value of is the total number of data points.

[0029] See also Figure 5 and Figure 2 , the decision-making ability assessment module includes: The decision point extraction submodule analyzes the decision behavior data based on the behavior pattern indicators, marks the operation location and timestamp of the decision point, and screens the key decision nodes that affect the overall decision. The execution process of the decision point set is as follows; The decision point extraction submodule extracts decision-related data from behavioral pattern indicators based on behavioral pattern indicators, with special attention to behavioral patterns related to major decisions, such as reaction time, diversity of choices, error rate, etc. The module applies algorithms to analyze the operation location, determine the geographical distribution and environmental context of decision points, and understand the decision background and influencing factors. Through timestamp data, the module analyzes the time series of decision-making and identifies temporal patterns or laws, such as the periodicity and frequency of decisions and their time-related changing trends. The submodule selects the key nodes that affect the overall decision, and uses influence analysis methods to evaluate the contribution of each decision point to the decision result. By calculating the correlation and causal strength between the decision point and the result, a set of decision points is obtained. The formula is: ; in, Indicates decision point The total influence of is the decision influence weight, It is a decision point In the The scores on the decision-making factors, is the total number of impact factors.

[0030] The efficiency comparison analysis submodule performs efficiency analysis on the decision points based on the decision point set, calculates the response time of the decision points, evaluates the processing speed of the decision points, compares the response time of multiple decision points, and obtains the execution process of the decision efficiency data as follows; The efficiency comparison analysis submodule is based on a set of decision points. It extracts the specific operation time of each decision point from the set of decision points, including the start and end time of the operation, as well as the time of each stage in the decision-making process. For the calculation of reaction time, the module defines the starting reaction time and decision completion time of each decision point, and uses timestamp data for accurate measurement. The total reaction time of each decision point is calculated, and the time is summarized and analyzed by statistical methods to identify patterns or anomalies in time. The submodule evaluates the processing speed of each decision point, which involves a detailed analysis of the time of each stage in the decision-making process, as well as calculating the average time required to process each decision, to obtain decision efficiency data. The formula is: ; in, Indicates decision point efficiency, It is a decision point The total reaction time, At the decision point The number of tasks that were successfully processed, It is a decision point The total number of tasks.

[0031] The quality assessment submodule uses decision tree analysis technology based on decision efficiency data to evaluate employees' response speed and selection quality, analyze the accuracy and consistency of decisions, compare the selection results of multiple decision points, and evaluate employees' performance in the decision-making process. The execution process of obtaining the decision quality index is as follows; Decision tree analysis technology follows the formula: ; in, For collection and The entropy of is the target dataset, is the set of decision points, For decision point The following subset, For the subset The probability of the i-th element in , is the balancing factor, It is a decision point The quality rating of and Represents the number of elements in the dataset and subset respectively, is the number of decision points, is the number of categories, is the number of rating dimensions.

[0032] The execution flow is as follows: Determine the total number of elements in each category in the data set S, and then for each decision point , determine the subset that influences this decision , calculate each subset The probability of each element appearing in , then calculate the weighted entropy sum of each subset and introduce the quality score and balance factor To adjust the calculation of the entropy so that it not only reflects the uncertainty of the probability distribution, but also reflects the quality of the decision point. By adjusting the value of the balance factor, the calculation of the entropy is optimized to adapt to different decision-making scenarios.

[0033] See also Figure 6 and Figure 2 , the learning habit recognition module includes: The decision frequency analysis submodule records the time intervals of employee decisions based on the decision quality index, marks the moment and frequency of decisions, and generates the decision frequency data through the following execution process; The decision frequency analysis submodule collects time data of decision events based on the decision quality index, including the specific time point of the decision and the total time required to complete the decision. The module evaluates the promptness and timeliness of employee decisions. The submodule monitors all decision-making behaviors of each employee in a specific period and records the exact timestamp of the decision. By analyzing the timestamps, the module calculates the frequency of decision-making, that is, the number of decisions per unit time, and the average time interval between two decisions. The analysis of time intervals helps to reveal the rhythm and continuity of employee decision-making. In order to further analyze the relationship between decision frequency and decision quality, the submodule scores each decision according to the quality index set in advance to generate decision frequency data. The formula is: ; in, represents the weighted decision frequency per unit time, It is The quality index of the decision, is the total observation time, is the number of decisions made during the observation period.

[0034] The decision complexity assessment submodule analyzes and quantifies the number of variables and the degree of interaction at key decision points based on decision frequency data, records the decision path, and tracks the execution time and action behaviors required for employees to make decisions. The execution process of generating decision complexity data is as follows; The decision complexity assessment submodule identifies and classifies the variables involved in each decision point in detail based on the decision frequency data to determine the structural complexity of the decision. The submodule collects all relevant data about the decision point, including the parameters involved in the decision, the selection options, and the consequences of the decision. The number of variables at each decision point directly affects the complexity of the decision; the more variables there are, the more information needs to be considered, and the more complex the decision process is. The module analyzes the degree of interaction between variables, such as whether there are dependencies between variables and how these relationships affect the decision process. The decision path is recorded and analyzed to track the complete process from the beginning to the end of the decision. It includes the information collection stage before the decision, the option evaluation stage, and the implementation stage of the final choice. Recording the path can reveal the logic and step sequence of the decision, further increasing the understanding of the complexity of the decision. In order to accurately measure the time cost and actions involved in each decision, the submodule records the start and end time points, as well as each action in the decision process, to generate decision complexity data. The formula is: ; in, represents the decision complexity, It is The influence weight of the decision variables, is the number of activities of the variable in the decision process, is the total execution time of the decision, is the total number of variables involved in the decision.

[0035] The learning behavior classification submodule identifies the decision-making patterns of employees based on decision complexity data, compares the deviations from the standard patterns, analyzes the differences in decision-making behaviors and selection paths, and classifies the learning behaviors of employees. The execution process of obtaining the learning style identification results is as follows; The learning behavior classification submodule is based on decision complexity data and uses data to understand how employees process information and make decisions. The submodule analyzes employee performance in different decision scenarios, recording the specific parameters of each decision, the steps taken, the time spent, and the final results. By comparing decision data with the standard decision model set by the company, the submodule is able to identify deviations and characteristics in employee decision-making behavior. Through a detailed analysis of decision paths, the module evaluates employees' choice preferences and strategy application in the decision-making process. This includes analyzing the variable selection, risk assessment, and priority setting involved in decision making. Different decision paths reveal how employees think and solve problems when faced with complex situations. The submodule applies machine learning classification algorithms, such as decision trees, random forests, or support vector machines, to classify employee behavior data into different learning style categories. The classification is based on the frequency, complexity, and efficiency of decisions demonstrated by employees in a simulated environment, as well as their performance under different types of decisions, to obtain learning style identification results.

[0036] See also Figure 7 and Figure 2 , the interactive mode mapping module includes: Based on the learning style identification results, the interaction record submodule extracts individual and group interaction data, collects employees' interaction records in the virtual environment, records the number of members' interactions and interaction objects, and generates the execution process of the interaction data set as follows; The interaction record submodule extracts individual and group interaction data based on the learning style identification results. The module analyzes the collaboration and communication patterns between team members and how the interaction is affected by their respective learning styles. The submodule tracks and records all interaction activities between employees in real time through the monitoring tools of the virtual environment. Including conversations, participation in collaborative tasks, use of shared resources, etc. The time of the interaction, the participants, and the content and results of the interaction are recorded for each interaction. The submodule analyzes the interaction data, calculates the number of interactions between each employee and other team members, and identifies the main interaction partners. It reveals the key communication nodes in the team and identifies the existing communication barriers or insufficient collaboration. The module analyzes the impact of each learning style on the interaction pattern based on the learning style identification results of the employees. Observe whether employees who tend to reflect on their learning style participate more in in-depth discussions, and whether employees who tend to practice their learning style participate more frequently in practical tasks. The formula is: ; in, represents the average interaction density, It is Number of interactions per employee, is the total number of team members.

[0037] The interaction frequency drawing submodule counts the interaction frequency of members based on the interaction data set, analyzes the time distribution of interaction events, records the interaction status in key time periods, and generates the interaction frequency graph. The execution process is as follows; The interaction frequency mapping submodule visualizes complex interaction data based on the interaction data set, allowing management to intuitively understand the interaction patterns and their time dynamics between team members. The submodule focuses on in-depth frequency analysis of the collected interaction data. Calculate the number of interactions for each employee within a specified time and compare the data with the team's average interaction frequency. Sort and classify the timestamps of each interaction to analyze the peak and trough of interaction during the day. The module performs statistical analysis on the time distribution of interaction events, including determining the time periods when interactions occur most frequently and the periods when there are relatively few interactions. The analysis results help identify the rhythm of team work and the time periods when communication and collaboration need to be enhanced. The module uses statistical data to generate an interaction frequency graph, which shows the changes in interaction frequency in different time periods and the interaction intensity between different team members to generate an interaction frequency graph. The formula is: ; in, Indicates The frequency of interaction between employees, is the number of employee interactions, is the length of the observation time window.

[0038] The team role analysis submodule analyzes the interaction relationship of employees in the virtual environment based on the interaction frequency graph, assigns team roles, and analyzes the influence and role distribution among members by evaluating the number and breadth of interactions. The execution process of generating team interaction effectiveness is as follows; The team role analysis submodule is based on the interaction frequency graph. The submodule conducts in-depth analysis of the interaction frequency graph to identify the employees who interact most frequently and the key individuals who serve as communication hubs. By calculating the number of interactions of each employee and the breadth of their interactions with different team members, the module evaluates the individual's social network position and potential influence in the team. The submodule uses data to analyze and define various roles in the team, such as leaders, intermediaries, knowledge sharers, etc. The allocation of roles is based on the employee's interaction pattern and behavioral characteristics in the team. The team interaction effectiveness is generated based on the frequency and scope of interaction and the contribution to the team's goals. The formula is: ; in, Indicates The role weight of each employee, Indicates Employees and The intensity of interaction between employees, It is The weight of an employee in the team, is the total number of team members.

[0039] See also Figure 8 and Figure 2 , the employee capability difference analysis module includes: The employee classification submodule is based on team interaction effectiveness. According to the performance of employees in the simulation environment, employees are classified into management personnel, technical personnel and grassroots employees according to performance data. The classification threshold is set by statistical analysis, and the execution process of generating the employee category list is as follows; The employee classification submodule is based on team interaction effectiveness. Based on the obtained team interaction data, such as the interaction frequency, interaction quality and decision-making efficiency between members, the module cleans and standardizes the data to ensure data quality and consistency, and calculates the comprehensive interaction index of each employee. The comprehensive index consists of three parts: interaction frequency score, interaction quality score and decision-making efficiency score. Each score is directly calculated from the corresponding data. The comprehensive interaction index is then compared with the preset classification threshold. The classification threshold is determined through historical data analysis and expert evaluation. The comprehensive index of each employee is ranked and the employees are divided into three categories according to the set threshold: managers, technicians and grassroots employees. The threshold adjustment method used in the classification process is a dynamic adjustment strategy based on historical quantiles to adapt to changes in different time periods and team conditions. This strategy ensures the accuracy and adaptability of the classification and generates a list of employee categories.

[0040] The skill assessment submodule simulates the target responsibility scenarios of multiple types of employees based on the employee category list, and verifies the professional skills of multiple employees through real-time operations, including problem solving, decision making and customer service capabilities. The execution process for generating skill analysis results is as follows; The skill assessment submodule is based on the employee category list. The module designs targeted duty scenarios based on the employee category list and creates different simulation environments for managers, technicians and grassroots employees. Each environment is designed to examine the key skills of employees in the corresponding category. High-pressure decision-making tasks are set for managers, troubleshooting tasks are set for technicians, and customer interaction simulations are set for grassroots employees. The performance of employees is recorded through real-time data of simulated operations, such as task completion time, error rate, and customer feedback scores. Through quantitative data collection, a comprehensive score is given in combination with the difficulty coefficient of the task and environmental variables. A standardized scoring model is used to calculate the skill index of each employee. The model takes into account the nature and influencing factors of each task to ensure the consistency and comparability of the score and generate the skill analysis results. The formula is: ; in, Represents the overall skill index of employees, It is The score of the task, is the weight coefficient of the task.

[0041] The strategy formulation submodule formulates exclusive training strategies based on the results of skill analysis, according to the classification results and professional skills of employees, including setting training goals, selecting training methods, arranging training schedules, adjusting training content through training effect feedback loops, and generating the execution process of customized training plans for positions as follows; The strategy formulation submodule is based on the skill analysis results. The module analyzes the skill analysis results from the skill assessment submodule, conducts in-depth analysis of the skill differences and needs of each employee category, and sets corresponding training goals based on the specific skill shortcomings and strengths of managers, technicians, and grassroots employees, such as improving decision-making efficiency, enhancing technical problem-solving capabilities, or optimizing customer service processes. It selects training methods suitable for each goal, such as simulation training, online courses, or on-site seminars, and formulates a specific training schedule. Considering the work arrangements of employees and the availability of training resources, it arranges reasonable training time and frequency. By setting up an initial feedback collection mechanism, such as post-training skill tests and employee self-evaluation, it collects training effect data, and dynamically adjusts the training content and methods based on data feedback to maximize the effect and efficiency of training, and generates a customized training plan for each position. The formula is: ; in, Represents the average improvement index of training effect, and Represent the skill assessment scores before and after training, The number of employees who participated in the training.

[0042] See also Fig. 9 and Figure 2 , the training effect verification module includes: The team collaboration task submodule customizes training programs based on positions, simulates team collaboration tasks in virtual reality, assigns roles to team members, defines the responsibilities and tasks of roles, establishes team collaboration task scenarios, and generates team task data. The execution process is as follows; The team collaboration task submodule customizes training programs based on positions. The submodule uses the team interaction effectiveness information obtained from previous analysis, including the frequency of interaction between members, role distribution and individual influence, to reasonably allocate roles based on this. Each team member is assigned corresponding roles and responsibilities based on their position in the team and the capabilities they have demonstrated before, such as project leader, coordinator, executor, etc. The submodule defines the specific responsibilities of each role and the tasks expected to be completed. Responsibilities and tasks are tailored to the team's goals and the characteristics of each role to ensure that each member can exert the greatest influence in the areas in which they are best. Including detailed descriptions of specific action guidelines and expected results for each role in collaborative tasks. Design and establish team collaboration task scenarios in virtual reality. The scenarios simulate the challenges encountered in real work environments, such as time pressure, resource constraints and complex problem-solving needs. The design of the scenarios promotes communication, decision-making and joint execution of tasks among team members, and generates team task data. The formula is: ; in, Indicates the efficiency of collaboration. It is Quality rating of interactions, is the frequency of the corresponding interaction, is the time required to complete the associated task, is the number of interactions recorded.

[0043] The real-time monitoring submodule monitors the collaboration and task completion of employees in real-time operations based on team task data, records members’ task progress and interactive behaviors, records employees’ abnormal operations and error times, and analyzes the number of collaborations and interaction frequencies. The execution process for generating real-time operation records is as follows; The real-time monitoring submodule is based on team task data. The module tracks every step of the employees in real time through an integrated tracking system, including their actions on specific tasks, the tools used, the time of operation, and the interaction with other team members. Data is collected in real time through sensing and data acquisition technology and updated to the system database in real time. The module records the task progress of each member in detail, including the start of the task, the various stages in progress, and the time of completion. The module also records all interactive behaviors related to the task, such as the frequency of collaborative discussions, the quality of execution of collaborative tasks, and any issues that require team coordination to resolve. The module closely monitors and records any abnormal operations or errors, including unplanned operations, deviations from preset processes, or operational errors. Records help to correct problems in a timely manner and prevent future errors, while providing a basis for training and improving work processes, and generating real-time operation records. The formula is: ; in, represents the average collaboration frequency, It is The number of interactions, is the total observation time, is the total number of interaction events.

[0044] The collaboration impact analysis submodule uses the random forest algorithm based on real-time operation records to analyze the impact of team collaboration on individual learning and overall performance, evaluate members’ performance in tasks, and compare individual and team task completion. By analyzing the optimization of learning effects and changes in the overall team effectiveness caused by collaboration, the execution process of obtaining the training application evaluation results is as follows; The random forest algorithm follows the formula: ; in, To predict the results, is the number of decision trees, For the The prediction function of a decision tree, is the input feature, For the The set of random parameters associated with each decision tree, For the The weight of a decision tree, is the control parameter of the decision tree, is the minimum number of samples for splitting a decision tree node, is the balance coefficient, is the regularization parameter.

[0045] The execution flow is as follows: Determining the number of decision trees , assigning a random parameter set to each tree , depth control parameters The minimum number of samples for node splitting , perform weighted average of the output of each tree, where the weight The performance of the tree is determined by the validation data set to adjust the influence of each tree in the final prediction and add a balance coefficient and the regularization parameter To optimize the model and prevent overfitting.

[0046] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A job competency evaluation system based on virtual reality technology, characterized in that: The system comprises: The behavior data capture module records employee behavior data including location movement, operation frequency, and decision-making duration based on the employee behavior data obtained by the VR device, verifies the integrity and timeliness of the data, and obtains a set of behavior characteristics; The behavior dynamic analysis module screens the key time points that affect employee behavior based on the behavior feature set, calculates the behavior variables at the time points, analyzes the behavior trends and abnormal patterns, and obtains the behavior pattern indicators; The decision-making ability evaluation module performs efficiency and result comparison analysis on the extracted decision points based on the behavior pattern indicators, evaluates the employee's response speed and selection quality, and obtains a decision-making quality index; The learning habit identification module analyzes the decision frequency and complexity of employees based on the decision quality index, identifies the decision pattern, compares the deviation from the standard pattern, and obtains the learning style identification result; The interaction mode mapping module extracts the interaction data of individuals and groups based on the learning style identification results, draws a frequency graph of member interactions, and obtains the team interaction effectiveness; The employee capability difference analysis module records the interaction frequency and influence of employees based on the team interaction effectiveness, quantifies the collaboration and leadership capabilities of employees, simulates the work scenarios of target responsibilities, evaluates the professional skills and strategy application of employees in various positions, and generates customized training programs for each position; The training effect verification module sets team tasks in virtual reality based on the team interaction effectiveness, monitors the completion of tasks, evaluates the impact of teamwork on individual and collective performance, and obtains training application evaluation results.

2. The job competency evaluation system based on virtual reality technology according to claim 1 is characterized by: The behavioral feature set includes employee location coordinates, operation timestamps and decision response time; the behavioral pattern indicators include time-sensitive behavior points, speed changes, and path deviation indicators; the decision quality index includes decision speed index, decision accuracy index and decision consistency score; the learning style identification results include decision frequency measurement, problem solving complexity rating and standard model deviation comparison data; the team interaction effectiveness includes team member interaction frequency graph, role adaptability analysis results and teamwork efficiency evaluation results; the job customized training program includes management skills optimization, technical problem solution, and customer interaction efficiency optimization; the training application evaluation results include task completion time records, team task completion, and individual performance data.

3. The job competency evaluation system based on virtual reality technology according to claim 1 is characterized in that: The behavior data capturing module comprises: The behavior monitoring submodule monitors the dynamic position of employees in the virtual environment in real time based on the employee behavior data obtained by the VR device, records the real-time reaction speed of employee operations, captures the key time points of decision-making events, and generates an initialization behavior data set; The data synchronization submodule performs standardization processing on the data based on the initialization behavior data set, converts the data format uniformly, eliminates errors and redundancies in the data, and generates formatted behavior data; The data verification submodule detects the integrity and timeliness of the data based on the formatted behavior data, updates the timestamp of the behavior information in real time, serializes and marks the data blocks, and generates a behavior feature set.

4. The job competency evaluation system based on virtual reality technology according to claim 1 is characterized in that: The behavior dynamics analysis module includes: The behavior time point screening submodule screens the key time points that affect the employee's behavior based on the behavior feature set, matches the timestamp data with the behavior event, and analyzes the key time period according to the employee's position change and operation frequency in the VR environment to obtain a key time point set; The behavior variable calculation submodule calculates the behavior variables at the time points based on the key time point set, calculates the speed changes between various time points through time series, calculates the path deviation range through trajectory deviation, compares the real-time path with the preset path, and generates a behavior variable set; The behavior pattern analysis submodule performs cluster analysis of the behavior patterns based on the set of behavior variables, groups the behavior data, extracts the characteristic vectors of the behavior variables, analyzes the fit and differences between the variables, identifies the pattern associations between the behavior variables, detects abnormal behavior characteristics, and obtains behavior pattern indicators.

5. The job competency evaluation system based on virtual reality technology according to claim 1 is characterized in that: The decision-making ability assessment module includes: The decision point extraction submodule analyzes the decision behavior data based on the behavior pattern indicators, marks the operation location and timestamp of the decision points, screens the key decision nodes that affect the overall decision, and obtains a decision point set; The efficiency comparison and analysis submodule performs efficiency analysis on the decision points based on the decision point set, calculates the reaction time of the decision points, evaluates the processing speed of the decision points, compares the reaction time of multiple decision points, and obtains decision efficiency data; The quality assessment submodule uses decision tree analysis technology based on the decision efficiency data to evaluate the employee's response speed and selection quality, analyze the accuracy and consistency of the decision, compare the selection results of multiple decision points, evaluate the employee's performance in the decision-making process, and obtain the decision quality index; The decision tree analysis technique follows the formula: ; in, For collection and The entropy of is the target dataset, is the set of decision points, For decision point The following subset, For the subset The probability of the i-th element in , is the balancing factor, It is a decision point The quality rating of and Represents the number of elements in the dataset and subset respectively, is the number of decision points, is the number of categories, is the number of rating dimensions.

6. The job competency evaluation system based on virtual reality technology according to claim 1 is characterized in that: The learning habit recognition module includes: The decision frequency analysis submodule records the time intervals of employee decisions based on the decision quality index, marks the time and frequency of decisions, and generates decision frequency data; The decision complexity assessment submodule analyzes and quantifies the number of variables and the degree of interaction at key decision points based on the decision frequency data, records the decision path, tracks the execution time and action behaviors required for employees to make decisions, and generates decision complexity data; The learning behavior classification submodule identifies the employee's decision-making pattern based on the decision complexity data, compares the deviation from the standard pattern, analyzes the difference in decision-making behavior and selection path, classifies the employee's learning behavior, and obtains the learning style identification result.

7. The job competency evaluation system based on virtual reality technology according to claim 1 is characterized in that: The interactive mode mapping module includes: The interaction record submodule extracts individual and group interaction data based on the learning style identification results, collects employee interaction records in the virtual environment, records the number of members' interactions and interaction objects, and generates an interaction data set; The interaction frequency drawing submodule counts the interaction frequency of members based on the interaction data set, analyzes the time distribution of interaction events, records the interaction status in key time periods, and generates an interaction frequency graph; The team role analysis submodule analyzes the interaction relationship of employees in the virtual environment based on the interaction frequency graph, assigns team roles, and generates team interaction effectiveness by evaluating the number and breadth of interactions, analyzing the influence and role distribution among members.

8. The job competency evaluation system based on virtual reality technology according to claim 1 is characterized in that: The employee capability difference analysis module includes: The employee classification submodule classifies employees into management personnel, technical personnel and grassroots employees according to the performance data based on the team interaction effectiveness and the performance of the employees in the simulation environment, and uses statistical analysis to set classification thresholds to generate an employee category list; The skill assessment submodule simulates target responsibility scenarios of multiple types of employees based on the employee category list, verifies the professional skills of multiple types of employees through real-time operations, including problem solving, decision making and customer service capabilities, and generates skill analysis results; The strategy formulation submodule formulates exclusive training strategies based on the skill analysis results and the employee's classification results and professional skills, including setting training goals, selecting training methods, arranging training schedules, adjusting training content through a training effect feedback loop, and generating job-customized training plans.

9. The job competency evaluation system based on virtual reality technology according to claim 1 is characterized in that: The training effect verification module includes: The team collaboration task submodule customizes the training program based on the positions, simulates the team collaboration tasks in virtual reality, assigns roles to team members, defines the responsibilities and tasks of the roles, establishes team collaboration task scenarios, and generates team task data; The real-time monitoring submodule monitors the collaboration and task completion of employees in real-time operations based on the team task data, records the task progress and interactive behaviors of members, records the number of abnormal operations and errors of employees, and analyzes the number of collaborations and interaction frequencies to generate real-time operation records; The collaboration impact analysis submodule uses a random forest algorithm based on the real-time operation records to analyze the impact of team collaboration on individual learning and overall performance, evaluate the performance of members in the task, compare the task completion of individuals and teams, and obtain the training application evaluation results by analyzing the optimization of learning effects and changes in the overall team effectiveness caused by collaboration.

10. The job competency evaluation system based on virtual reality technology according to claim 9 is characterized in that: The random forest algorithm follows the formula: ; in, To predict the results, is the number of decision trees, For the The prediction function of a decision tree, is the input feature, For the The set of random parameters associated with each decision tree, For the The weight of a decision tree, is the control parameter of the decision tree, is the minimum number of samples for splitting a decision tree node, is the balance coefficient, is the regularization parameter.

Citation Information

Cited By

  • Preschool child personalized learning path planning method and system of online education system

    CN119648494A

  • Cancer survivor post returning support system

    CN119991061A

  • Employment matching method and system based on intelligent data analysis

    CN120218560A

  • Dynamic behavior style test and post adaptation degree screening evaluation platform and evaluation method

    CN120297929A

  • Virtual workplace multi-dimensional assessment method and system based on deep reinforcement learning

    CN120373970A