A method for safety management and on-the-job performance management of external personnel in the military industry

Abstract

This invention discloses a method for security management and on-the-job performance management of external personnel in the military industry, relating to the field of access management technology. The method includes: acquiring the identity information, job information, visited area information, and electronic device information of visiting personnel; generating path information based on the visited area information and acquiring electronic restraint information; using the path information and electronic restraint information for security management of external personnel; granting access permissions to external personnel upon arrival at the visited area within a preset passage time; tracking the business workflow of external personnel during their work process; extracting performance evaluation indicators and constructing a performance evaluation model based on performance time; and conducting performance evaluation of external personnel's security responsibility system. This invention improves the efficiency and security of external personnel access management through information technology, and manages the performance of external personnel, generating visualized data to standardize the performance behavior of operational personnel.

Description

Technical Field

[0001] This invention relates to the field of access management technology, and more specifically, to a method for security management and on-duty performance management of external personnel in the military industry. Background Technology

[0002] Military enterprises have a large number of external units involved in their business and work, involving the entry of outsiders, including visitors, logistics personnel, temporary employees, etc., which will inevitably increase the complexity of enterprise management, increase the safety risks, and increase the difficulty of management. The difficulties in the safety management of outsiders are mainly concentrated in the following aspects: (1) The information of personnel is mixed with the truth and the data cannot be accurate. In the daily operation of military enterprises, they face the frequent flow of outside laborers and the messy personnel data. The collection of personnel information is often left to the personnel to fill in themselves. The enterprise does not have a corresponding database and cannot verify it. Therefore, the data is often mixed with the truth and cannot be refined and accurate. (2) The level of personnel is uneven and it is difficult to quickly identify them. Because of the frequent flow of personnel and the strong timeliness of various tasks, it is difficult to quickly identify the quality of personnel. (3) Skills and safety training is inadequate. Most companies conduct onboard training to popularize safety knowledge, master safe operating skills, and learn on-site operating procedures. However, most companies mainly rely on offline explanations or verbal communication, which only covers the training process but rarely achieves the training objectives or accurately assesses the training results and quality.

[0003] For visitors, effective management and monitoring are necessary during their visits to ensure the security and interests of military enterprises. Traditional manual management methods suffer from low efficiency and security, and are no longer able to meet rapidly changing needs. Therefore, how to achieve efficient security management and comprehensive on-duty performance management for visitors is an urgent problem to be solved. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention proposes a method for the safety management and on-the-job performance management of external personnel in the military industry. The aim is to improve the efficiency and security of access management for external personnel through information technology, and to manage their performance of duties, thereby generating visualized data to standardize the performance of personnel.

[0005] This invention provides a method for safety management and on-the-job performance management of external personnel in the military industry, comprising the following steps:

[0006] The system acquires the identity information, job information, visited area information, and electronic device information of visiting personnel; generates path information based on the visited area information; generates initial electronic restraint information based on the path information; and sends the path information and initial electronic restraint information to the visiting personnel's electronic devices.

[0007] The path information is divided into multiple sub-paths based on the travel area in the path information, a preset travel time for each sub-path is generated, and the electronic constraint information is updated based on the preset travel time.

[0008] Access is granted to visitors once they arrive at the area within a preset time limit. Based on the functional information, the business workflow of the visitors during their work is tracked, and performance evaluation indicators are extracted from the business workflow.

[0009] A performance evaluation model is constructed based on performance evaluation indicators and performance time, and multi-source performance data is obtained as model input to evaluate the performance of the safety responsibility system for external personnel.

[0010] This plan obtains the following information in the form of visiting visitors: their identity information, job information, visited area information, and electronic device information:

[0011] Collect the identity information of visitors and determine whether they are on the visitor blacklist. If they are on the visitor blacklist, adding visitors is not allowed. If they are not on the visitor blacklist, the identity information is used to determine whether the visitor has a visit appointment.

[0012] If the identity information exists, the visit appointment information is segmented according to the preset word segmentation template after the identity information is verified, and the job information, visited area information and visit purpose of the visitor are read. If the identity information does not exist, the visitor registration template is called to collect the job information, visited area information and visit purpose of the visitor.

[0013] When verifying the identity of visitors, facial recognition and electronic device information verification are performed. A link is established with the visitor's electronic device. After successful verification, the visitor and their corresponding multi-dimensional visit information are added to the visitor list.

[0014] In this scheme, path information is generated based on the visited area information, and initial electronic constraint information is generated based on the path information, specifically as follows:

[0015] After determining the area visited by visitors, select the starting point and ending point of visitors, generate path information based on the starting point and ending point, obtain the areas traversed by the path information, and extract the regional responsibility features of all traversed areas.

[0016] Sensitive words are preset based on the regional responsibility characteristics, and sensitive information filtering standards for different data modalities are constructed through the sensitive words. The sensitive information filtering standards are then matched with the regions traversed.

[0017] In addition, based on the regional responsibility characteristics and the pre-set confidentiality requirements of military enterprises, the areas traversed are classified into confidentiality levels. The confidentiality level is used to generate terminal restrictions in the regional strategy management. The sensitive information filtering standards and terminal restrictions of each traversed area are linked together according to the passage order of the path information to generate initial electronic constraint information.

[0018] In this scheme, the path information is divided into multiple sub-paths based on the travel area in the path information, a preset travel time for each sub-path is generated, and the electronic constraint information is updated based on the preset travel time. Specifically:

[0019] The path information of visitors is segmented according to the monitoring equipment in the area they travel through, and several sub-paths are obtained. The starting point and ending point of each sub-path are extracted to generate sub-path segments. The historical passage time of the personnel in the sub-path segments is extracted based on the historical monitoring data of the monitoring equipment.

[0020] The extracted historical passage time is averaged, and a baseline passage time is set for different sub-paths based on the average passage time. The facial information of visitors is captured by monitoring equipment and the location information of electronic devices is obtained. The passage speed of visitors is obtained based on facial information verification and changes in location information.

[0021] The baseline passage time is weighted based on the passage speed to generate weight information, and preset passage times for different sub-paths are generated. The electronic restraint information is updated using the preset passage times for each sub-path. When an outsider enters the fenced area of ​​a sub-path, the corresponding electronic restraint information is invoked to restrict the terminal.

[0022] In this plan, access permissions are granted to visitors once they arrive at the area within a preset access time. Specifically:

[0023] When an outsider enters any sub-path, the monitoring equipment obtains the outsider's facial information. After the facial verification is successful, the start timestamp is recorded, and the location information of the electronic device is recorded. When the deviation between the location information and the path information is greater than a preset distance, trajectory correction information is generated, and the outsider is reminded through the electronic device.

[0024] When the location information reaches the end of the sub-path, the facial information of the visitor is obtained again through the monitoring equipment. After the facial verification is passed, the passage time is obtained by subtracting the start time stamp from the end timestamp.

[0025] Determine whether the passage time of the visitor is greater than the preset passage time. If it is greater, extract the visitor's passage trajectory in the sub-path, extract the visitor's stop point based on the passage trajectory, analyze the stop location and stop time based on the extracted stop point, and display it visually.

[0026] The system obtains the number of times that the visitor's passage time exceeds the preset passage time. When the number of times is less than the preset threshold, it is considered that the visitor has arrived at the area within the preset passage time, and access permission is granted to the visitor.

[0027] In this solution, the workflow of external personnel during their work process is tracked based on the aforementioned functional information, and performance evaluation indicators are extracted through the workflow. Specifically:

[0028] Based on the functional information of the visitors, determine their list of responsibilities, extract business keywords using the list of responsibilities and the purpose of their visit, construct a retrieval task based on the business keywords, use the retrieval task to obtain business execution data that meets the retrieval requirements, and perform structured preprocessing on the business execution data;

[0029] The preprocessed business execution data is divided using the aforementioned list of responsibilities. Based on the divided business execution data, the business workflow of external personnel is populated, and a business dataset with responsibility business classification labels is constructed.

[0030] In the business dataset, the business execution data with the highest correlation to the performance evaluation is obtained based on the Pearson correlation coefficient. The remaining business execution data in the business dataset are introduced one by one according to the selected business data. The mRMR algorithm is used to obtain the importance score of the business execution data based on the principle of maximizing the correlation between the business execution data and the performance evaluation and minimizing the redundancy between the business execution data.

[0031] Based on the importance score, a preset number of business execution data are selected to generate performance evaluation indicators.

[0032] In this plan, a performance evaluation model is constructed based on performance evaluation indicators and performance time, specifically as follows:

[0033] Obtain the performance time of the business execution data corresponding to the performance evaluation indicators, match the performance evaluation indicators according to the performance time to construct performance evaluation rules, and construct a performance evaluation model based on autoencoder and multilayer perceptron.

[0034] The performance evaluation model learns performance evaluation rules through iterative training, uses an autoencoder as the input layer to encode the performance evaluation rules, imports the encoded performance evaluation rules into a multilayer perceptron for deep feature extraction, and outputs the evaluation results through a fully connected layer.

[0035] In the training process of the performance evaluation model, the particle swarm optimization algorithm is introduced to optimize the network structure. The evaluation accuracy of the performance evaluation model is used as the particle fitness. The position and velocity of the particles are continuously updated according to the particle fitness. When the termination condition is met, the number of layers of the autoencoder and the number of hidden layers of the multilayer perceptron are obtained according to the optimal position of the particles to establish the performance evaluation model.

[0036] In this plan, multi-source performance data is used as model input to evaluate the performance of external personnel in the safety responsibility system. Specifically:

[0037] Acquire multi-source performance data of external personnel, preprocess the multi-source performance data, and use it as the model input of the performance evaluation model. In the performance evaluation model, the multi-source performance data is encoded and embedded based on the performance evaluation indicators.

[0038] Based on the performance evaluation rules, the characteristic information of multi-source performance data is nonlinearly fitted to output the performance evaluation level of the safety responsibility system for migrant personnel.

[0039] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0040] This invention improves the efficiency and security of visitor access management through information technology. Visitor management based on big data technology enhances the accuracy and completeness of collected data, filling gaps in business data and addressing confidentiality requirements in the military industry. Automated end-to-end management reduces the need for managers to make assumptions or perform manual operations, significantly reducing their workload. Furthermore, the visualization of visitor performance data standardizes staff behavior, further improving management efficiency and accuracy. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments or examples of the present invention, the drawings used in the embodiments or examples will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained according to these drawings without creative effort.

[0042] Figure 1 A flowchart is shown for methods of safety management and on-the-job performance management for external personnel in the military industry;

[0043] Figure 2 A flowchart illustrating the management of visitor routes in this embodiment is shown.

[0044] Figure 3 The flowchart illustrating the extraction of performance evaluation indicators for external personnel in this embodiment is shown.

[0045] Figure 4 A block diagram of a security management and on-the-job performance management system for external personnel in the military industry is shown. Detailed Implementation

[0046] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0047] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0048] Figure 1 The flowcharts show the methods for safety management and on-the-job performance management of external personnel in the military industry.

[0049] like Figure 1 As shown, this embodiment provides a method for safety management and on-the-job performance management of external personnel in the military industry, including:

[0050] S102, obtain the identity information, job information, visited area information and electronic device information of the visiting personnel, generate path information based on the visited area information, generate initial electronic restraint information based on the path information, and send the path information and initial electronic restraint information to the electronic device of the visiting personnel;

[0051] S104, the path information is divided into multiple sub-paths according to the travel area in the path information, a preset travel time for the sub-paths is generated, and the electronic constraint information is updated according to the preset travel time.

[0052] S106: When an outsider arrives at the area within a preset passage time, access permission is granted. The business workflow of the outsider during the work process is tracked according to the functional information. Performance evaluation indicators are extracted through the business workflow.

[0053] S108. Based on the performance evaluation indicators and the performance time, a performance evaluation model is constructed, and multi-source performance data is obtained as the model input to conduct a performance evaluation of the safety responsibility system for external personnel.

[0054] It should be noted that using a visitor registration system or identity verification system to acquire and store basic information of temporary visitors can automatically obtain their basic information, saving human resources and time costs. The system automatically collects the identity information of visitors and determines whether they are on a visitor blacklist. If on the blacklist, adding visitors is not allowed; if not, the system uses the identity information to determine if the visitor has any appointment information. By setting up a visitor blacklist, security risks are reduced. If a visit appointment information exists, after verifying the identity information, the visit appointment information is segmented according to a preset word segmentation template to read the visitor's job information, visited area information, and visit purpose. For example, the job information of an equipment maintenance personnel is "repair," the visited area information is the area where the equipment to be repaired is located, and the visit purpose is "repair the equipment." If no such information exists, the visitor registration template is called to collect the visitor's job information, visited area information, and visit purpose. Facial information is correctly collected through a designated positioning hardware device to ensure a high accuracy rate of facial recognition. When verifying the identity of a visitor, facial verification and electronic device information verification are performed, and a link is established with the visitor's electronic device. After the device link is established, the relevant information of the electronic device will be managed in data visualization on the personnel management platform. After successful verification, the visitor and their corresponding multi-dimensional visit information are added to the visitor list.

[0055] The personnel management platform utilizes a visitor list for personnel management, supporting the creation, editing, moving, and deletion of visitor groups. It also supports displaying added visitor information in a list, including name, work unit, function, and electronic device. Visitor information can be searched by keywords, and the list header field can be sorted. Furthermore, applications, files, and policies can be distributed to selected visitors through a constructed workflow.

[0056] It should be noted that after determining the area visited by visitors, the starting point and ending point of visitors are selected. Path information is generated based on the starting point and ending point, the areas traversed by the path information are obtained, and the regional responsibility characteristics of all traversed areas are extracted. For example, the regional responsibility characteristic of the production workshop is military equipment production, and taking photos, recording audio, and recording video are prohibited in the production workshop area. Sensitive words are preset based on the regional responsibility characteristics. Sensitive information filtering standards for different data modalities are constructed using these sensitive words. The semantic information of the sensitive words is obtained. Image modal data and speech modal data are transferred and trained based on the semantic information of the sensitive words. Text, image, sentence, and speech analysis are performed to prevent unauthorized personnel from leaking secrets by recording or taking pictures through electronic devices. The sensitive information filtering standards are matched with the areas traversed. In addition, the confidentiality regulations of military enterprises are obtained. Confidentiality requirements are preset based on the confidentiality regulations and other documents. Based on the regional responsibility characteristics and the confidentiality requirements preset by military enterprises, the areas traversed are classified into confidentiality levels. Terminal restrictions in the area policy management are obtained from the confidentiality level, such as prohibiting photography at the first level, prohibiting photography and recording at the second level, etc. The sensitive information filtering standards and terminal restrictions of each traversed area are linked together according to the passage order of the path information to generate initial electronic constraint information.

[0057] Figure 2 A flowchart illustrating the management of visitor routes in an embodiment is shown.

[0058] According to an embodiment of the present invention, the path information is divided into multiple sub-paths based on the travel area in the path information, a preset travel time for each sub-path is generated, and the electronic constraint information is updated according to the preset travel time, specifically as follows:

[0059] S202, the path information of visitors is divided according to the monitoring equipment in the area they pass through, several sub-paths are obtained, the starting point and ending point of the sub-paths are extracted to generate sub-path segments, and the historical passage time of the personnel in the sub-path segments is extracted according to the historical monitoring data of the monitoring equipment.

[0060] S204, average the extracted historical passage time, set a baseline passage time for different sub-paths based on the average passage time, capture the facial information of visitors using monitoring equipment and obtain the location information of electronic devices, and obtain the passage speed of visitors based on facial information verification and location information changes;

[0061] S206, the baseline passage time is weighted according to the passage speed to generate weight information, and preset passage time for different sub-paths is generated. The electronic restraint information is updated using the preset passage time of each sub-path. When an outsider enters the fenced area of ​​a sub-path, the corresponding electronic restraint information is invoked to restrict the terminal.

[0062] It should be noted that when visitors enter any sub-path, their facial information is acquired through monitoring equipment. After facial verification is successful, the start timestamp is recorded, along with the location information of the electronic device. When the deviation between the location information and the path information exceeds a preset distance, trajectory correction information is generated and sent to the visitor's electronic device. The electronic device then alerts the visitor, ensuring that they always follow the preset path, thus guaranteeing visitor security and improving management efficiency. Furthermore, the passage time threshold for each sub-path is set individually based on the visitor's walking speed to avoid frequent warnings caused by the visitor's walking speed.

[0063] When the location information reaches the end of the sub-path, the facial information of the visitor is obtained again through the monitoring equipment. After the facial verification is successful, the passage time is obtained by subtracting the start time stamp from the end timestamp. It is then determined whether the visitor's passage time is greater than the preset passage time. If it is greater, the visitor's passage trajectory in the sub-path is extracted. Based on the passage trajectory, the visitor's stop points are extracted. Based on the extracted stop points, the stop location and stop time are analyzed and displayed visually on the personnel management platform. The personnel management platform supports trajectory analysis, analysis of the start, end, and stop points of the sub-path, analysis of stop location, stop time, movement time, and movement distance, and determines the reason for the visitor's stay. When an anomaly occurs, the management personnel are alerted to conduct on-site investigation.

[0064] The system records the number of times a visitor's access duration exceeds a preset access duration. When this number is less than a preset threshold, the visitor is considered to have entered the area within the preset access duration, and access is granted. The system also performs segmented real-time monitoring of the visitor's route, recording the number of times the access duration is outside the preset duration. Based on whether the records meet preset rules, the system determines whether to close the temporary visitor's access. Furthermore, a blacklist threshold is set; when the recorded number exceeds the preset blacklist threshold, the visitor is added to the access blacklist.

[0065] Figure 3 The flowchart illustrating the extraction of performance evaluation indicators for external personnel in the embodiment is shown.

[0066] According to an embodiment of the present invention, the business workflow of external personnel during their work process is tracked based on the functional information, and performance evaluation indicators are extracted through the business workflow, specifically as follows:

[0067] S302, determine the list of responsibilities of the visitors based on their functional information, extract business keywords using the list of responsibilities and the purpose of their visit, construct a retrieval task based on the business keywords, obtain business execution data that meets the retrieval requirements using the retrieval task, and perform structured preprocessing on the business execution data;

[0068] S304, the preprocessed business execution data is divided according to the responsibility list, and the business workflow of external personnel is filled based on the divided business execution data to construct a business dataset with responsibility business classification labels;

[0069] S306, in the business dataset, the business execution data with the highest correlation to the performance evaluation is obtained according to the Pearson correlation coefficient. The remaining business execution data in the business dataset are introduced one by one according to the selected business data. The mRMR algorithm is used to obtain the importance score of the business execution data based on the principle of the maximum correlation between the business execution data and the performance evaluation and the minimum redundancy between the business execution data.

[0070] S308, Select a preset number of business execution data based on the importance score to generate performance evaluation indicators.

[0071] It should be noted that business keywords, such as maintenance process, equipment operation and maintenance, risk control, hazard management, and safety inspection data, are extracted using the duty list and the purpose of the visit. The business execution data undergoes structured preprocessing, including data cleaning, outlier removal, missing value filling, and clustering. The mRMR algorithm is used to select variables with low redundancy between features and high correlation between features and the dependent variable, which has advantages such as fast calculation speed and good stability. An improved mRMR algorithm based on the Pearson correlation coefficient is used to calculate the redundancy between business execution data and the correlation between business execution data and performance evaluation. The longest path in the business process is extracted from the time-series business execution data sequence to extract the main process path. Business execution process paths with the same task information and the same predecessor task in the business execution data are then identified based on the business nodes in the main process path. These business execution process paths are matched to obtain a candidate set of extended paths, which are selected according to business needs. The business workflow of external personnel is then obtained based on the selected main process path and extended paths.

[0072] The personnel management platform sets different attendance rules for different personnel and records detailed information on personnel entering and leaving the area. A performance evaluation model is constructed based on performance evaluation indicators and performance time. The platform obtains the performance time corresponding to the performance evaluation indicators, such as attendance time and equipment maintenance time. Performance evaluation rules are constructed by matching the performance time with the performance evaluation indicators. Business execution data for each performance evaluation indicator is collected, and the business execution data is scored to obtain the performance rate and corresponding score for each responsibility category. For example, a performance rate of 80% for maintenance on-duty time corresponds to a score of 80. A qualitative evaluation system is formed by setting graded levels according to the performance evaluation indicator scores, with four levels (A, B, C, and D): A represents excellent performance, B represents good performance, C represents average performance, and D represents poor performance. A performance evaluation model is constructed based on autoencoders and multilayer perceptrons. Multiple autoencoders with identical structures are sequentially connected and trained layer-by-layer in an unsupervised manner. Training sample data corresponding to business execution data is used for encoding and reconstruction, and the hidden layers are trained layer by layer. The performance evaluation model learns performance evaluation rules through iterative training. The autoencoders are used as input layers to encode the performance evaluation rules, which are then imported into the multilayer perceptron for deep feature extraction. The evaluation results are output through fully connected layers. During the training process, a particle swarm optimization algorithm is introduced to optimize the network structure. The evaluation accuracy of the performance evaluation model is used as the particle fitness. The position and velocity of the particles are continuously updated based on the particle fitness. When a termination condition is met, the number of autoencoder layers and the number of hidden layers in the multilayer perceptron are obtained based on the optimal particle position, thus establishing the performance evaluation model.

[0073] It should be noted that multi-source performance data of external personnel is obtained by utilizing a personnel management platform to acquire data such as pre-defined area fence check-in data, real-time video collection of on-duty time data, meeting attendance statistics, risk control data, hazard management data, and safety patrol data. This multi-source performance data is preprocessed and used as input to a performance evaluation model. The model encodes and embeds this multi-source performance data based on performance evaluation indicators. Real-time monitoring of external personnel's business execution data and performance quantitative score data is conducted. Based on pre-defined performance rules, nonlinear fitting is performed on the feature information of the multi-source performance data, and the performance quantitative score data is analyzed in real time to output the performance evaluation level of external personnel's safety responsibility system. Statistical analysis methods and data mining algorithms are used to establish feature labels for the external personnel's performance evaluation level, creating a performance profile model. This model is then visualized to improve the initiative and self-awareness of external personnel.

[0074] Figure 4The diagram shows a block diagram of the present invention for the safety management and on-the-job performance management system for external personnel in the military industry.

[0075] The second embodiment of the present invention provides a management system 4 for the safety management and on-duty performance management of external personnel in the military industry. The system includes the following units: a memory 41 and a processor 42. The memory and the processor store and execute a method program for the safety management and on-duty performance management of external personnel in the military industry.

[0076] The third embodiment of the present invention provides a computer-readable storage medium, which includes a program for a method of security management and on-the-job performance management of external personnel in the military industry. When the program for security management and on-the-job performance management of external personnel in the military industry is executed by a processor, it implements the steps of the above-mentioned method for security management and on-the-job performance management of external personnel in the military industry.

[0077] In the several embodiments provided in this application, it should be understood that the disclosed methods can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components may be combined, or integrated into another system, or some features may be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed may be through some interfaces, indirect coupling or communication connection of devices or units, and may be electrical, mechanical, or other forms.

[0078] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0079] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0080] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

[0081] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for safety management and on-the-job performance management of external personnel in the military industry, characterized in that, Includes the following steps: The system acquires the identity information, job information, visited area information, and electronic device information of visiting personnel; generates path information based on the visited area information; generates initial electronic restraint information based on the path information; and sends the path information and initial electronic restraint information to the visiting personnel's electronic devices. The path information is divided into multiple sub-paths based on the travel area in the path information, a preset travel time for each sub-path is generated, and the electronic constraint information is updated based on the preset travel time. Access is granted to visitors once they arrive at the area within a preset time limit. Based on the functional information, the business workflow of the visitors during their work is tracked, and performance evaluation indicators are extracted from the business workflow. A performance evaluation model is constructed based on performance evaluation indicators and performance time, and multi-source performance data is obtained as model input to evaluate the performance of the safety responsibility system for external personnel. Based on the aforementioned functional information, the workflow of external personnel during their work process is tracked, and performance evaluation indicators are extracted from the workflow, specifically: Based on the functional information of the visitors, determine their list of responsibilities, extract business keywords using the list of responsibilities and the purpose of their visit, construct a retrieval task based on the business keywords, use the retrieval task to obtain business execution data that meets the retrieval requirements, and perform structured preprocessing on the business execution data; The preprocessed business execution data is divided using the aforementioned list of responsibilities. Based on the divided business execution data, the business workflow of external personnel is populated, and a business dataset with responsibility business classification labels is constructed. In the business dataset, the business execution data with the highest correlation to the performance evaluation is obtained based on the Pearson correlation coefficient. The remaining business execution data in the business dataset are introduced one by one according to the selected business data. The mRMR algorithm is used to obtain the importance score of the business execution data based on the principle of maximizing the correlation between the business execution data and the performance evaluation and minimizing the redundancy between the business execution data. Based on the importance score, a preset number of business execution data are selected to generate performance evaluation indicators; A performance evaluation model is constructed based on performance evaluation indicators and performance duration, specifically as follows: Obtain the performance time of the business execution data corresponding to the performance evaluation indicators, match the performance evaluation indicators according to the performance time to construct performance evaluation rules, and construct a performance evaluation model based on autoencoder and multilayer perceptron. The performance evaluation model learns performance evaluation rules through iterative training, uses an autoencoder as the input layer to encode the performance evaluation rules, imports the encoded performance evaluation rules into a multilayer perceptron for deep feature extraction, and outputs the evaluation results through a fully connected layer. In the training process of the performance evaluation model, the particle swarm optimization algorithm is introduced to optimize the network structure. The evaluation accuracy of the performance evaluation model is used as the particle fitness. The position and velocity of the particles are continuously updated according to the particle fitness. When the termination condition is met, the number of layers of the autoencoder and the number of hidden layers of the multilayer perceptron are obtained according to the optimal position of the particles to establish the performance evaluation model. Multi-source performance data is used as model input to evaluate the performance of external personnel in the safety responsibility system. Specifically: Acquire multi-source performance data of external personnel, preprocess the multi-source performance data, and use it as the model input of the performance evaluation model. In the performance evaluation model, the multi-source performance data is encoded and embedded based on the performance evaluation indicators. Based on the performance evaluation rules, the characteristic information of multi-source performance data is nonlinearly fitted to output the performance evaluation level of the safety responsibility system for migrant workers.

2. The method for safety management and on-the-job performance management of external personnel in the military industry according to claim 1, characterized in that, Obtain the identity information, job information, visited area information, and electronic device information of visiting personnel, specifically: Collect the identity information of visitors and determine whether they are on the visitor blacklist. If they are on the visitor blacklist, adding visitors is not allowed. If they are not on the visitor blacklist, the identity information is used to determine whether the visitor has a visit appointment. If the identity information exists, the visit appointment information is segmented according to the preset word segmentation template after the identity information is verified, and the job information, visited area information and visit purpose of the visitor are read. If the identity information does not exist, the visitor registration template is called to collect the job information, visited area information and visit purpose of the visitor. When verifying the identity of visitors, facial recognition and electronic device information verification are performed. A link is established with the visitor's electronic device. After successful verification, the visitor and their corresponding multi-dimensional visit information are added to the visitor list.

3. The method for safety management and on-the-job performance management of external personnel in the military industry according to claim 1, characterized in that, Based on the visited area information, path information is generated, and based on the path information, initial electronic constraint information is generated, specifically as follows: After determining the area visited by visitors, select the starting point and ending point of visitors, generate path information based on the starting point and ending point, obtain the areas traversed by the path information, and extract the regional responsibility features of all traversed areas. Sensitive words are preset based on the regional responsibility characteristics, and sensitive information filtering standards for different data modalities are constructed through the sensitive words. The sensitive information filtering standards are then matched with the regions traversed. In addition, based on the regional responsibility characteristics and the pre-set confidentiality requirements of military enterprises, the areas traversed are classified into confidentiality levels. The confidentiality level is used to generate terminal restrictions in the regional strategy management. The sensitive information filtering standards and terminal restrictions of each traversed area are linked together according to the passage order of the path information to generate initial electronic constraint information.

4. A method for safety management and on-the-job performance management of external personnel in the military industry according to claim 1, characterized in that, The path information is divided into multiple sub-paths based on the travel area in the path information, a preset travel time for each sub-path is generated, and the electronic constraint information is updated based on the preset travel time, specifically as follows: The path information of visitors is segmented according to the monitoring equipment in the area they travel through, and several sub-paths are obtained. The starting point and ending point of each sub-path are extracted to generate sub-path segments. The historical passage time of the personnel in the sub-path segments is extracted based on the historical monitoring data of the monitoring equipment. The extracted historical passage time is averaged, and a baseline passage time is set for different sub-paths based on the average passage time. The facial information of visitors is captured by monitoring equipment and the location information of electronic devices is obtained. The passage speed of visitors is obtained based on facial information verification and changes in location information. The baseline passage time is weighted based on the passage speed to generate weight information, and preset passage times for different sub-paths are generated. The electronic restraint information is updated using the preset passage times for each sub-path. When an outsider enters the fenced area of ​​a sub-path, the corresponding electronic restraint information is invoked to restrict the terminal.

5. A method for safety management and on-the-job performance management of external personnel in the military industry according to claim 1, characterized in that, Access is granted to visitors once they arrive at the designated area within the preset time limit. Specifically: When an outsider enters any sub-path, the monitoring equipment obtains the outsider's facial information. After the facial verification is successful, the start timestamp is recorded, and the location information of the electronic device is recorded. When the deviation between the location information and the path information is greater than a preset distance, trajectory correction information is generated, and the outsider is reminded through the electronic device. When the location information reaches the end of the sub-path, the facial information of the visitor is obtained again through the monitoring equipment. After the facial verification is passed, the passage time is obtained by subtracting the start time stamp from the end timestamp. Determine whether the passage time of the visitor is greater than the preset passage time. If it is greater, extract the visitor's passage trajectory in the sub-path, extract the visitor's stop point based on the passage trajectory, analyze the stop location and stop time based on the extracted stop point, and display it visually. The system obtains the number of times that the visitor's passage time exceeds the preset passage time. When the number of times is less than the preset threshold, it is considered that the visitor has arrived at the area within the preset passage time, and access permission is granted to the visitor.

Images