A big data-based multi-platform recruitment information management system and method

Abstract

The application discloses a multi-platform recruitment information management system and method based on big data, relates to the technical field of recruitment information management, and effectively reflects production operation states through indexes such as sorting speed and calibration response, and replaces written certificates with AR simulation tests to realize more intuitive and operable talent qualification certification, and make up for the short board of information isolation, subjective evaluation and untimely verification in traditional certification; the dynamic capability evaluation model and the elastic team formation strategy generator work cooperatively, match post skills according to actual production requirements, dynamically adjust personnel combination ratios, and give consideration to cost and production progress, so that the problems of fragmented recruitment information, data lag and skill evaluation deviation existing in the background are effectively solved, and the scientificity and accuracy of recruitment and training decisions are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of recruitment information management, and in particular to a multi-platform recruitment information management system and method based on big data. BACKGROUND

[0002] In the current recruitment management system, when encountering high concurrency, strong professionalism and short cycle of labor demand, the traditional screening mechanism based on keyword matching will lead to a large number of mis-screening of workers with practical experience but without certificates due to excessive reliance on rigid thresholds of certificate authentication, such as setting HACCP certificate as a hard condition. The deeper contradiction of this phenomenon lies in the fragmentation of data value. The behavior data accumulated by each recruitment platform and the internal production data of enterprises, such as historical labor quality evaluation, are in an isolated state. The recruitment system can accurately match the education background, certificate and other explicit labels, but cannot screen the high-value information of the candidates in some aspects, resulting in that the recruited certificate holders cannot actually adapt to the actual production line rhythm.

[0003] This distortion of the ability evaluation system directly triggers a chain reaction. In order to fill the personnel gap, enterprises have to adopt high-priced recruitment and lower-standard recruitment strategies, falling into a vicious cycle, and the hidden loss caused by temporary training will cause disturbance to the production line;

[0004] As can be seen, the current recruitment management scheme mostly simplifies the value of talents as a combination of several static labels, ignoring the disassemblability of skill elements, the transferability of experience and the dynamic adaptability of person-job matching. When a sudden demand occurs, such a scheme cannot effectively identify the real ability, nor can it predict potential risks through data fusion. Ultimately, enterprises are forced to make difficult choices under the triple pressures of high cost, low efficiency and high risk. This systematic failure is particularly prominent in strong volatility industries such as fresh food processing and cross-border e-commerce, and has become a key bottleneck restricting the agile response of enterprises to the market. SUMMARY

[0005] In view of the above existing problems, the present application is proposed.

[0006] The present application provides a multi-platform recruitment information management system and method based on big data, which solves the problem that in the current multi-platform recruitment information management scheme, the value of talents is mostly simplified as a combination of several static labels, ignoring the disassemblability of skill elements, the transferability of experience and the dynamic adaptability of person-job matching. When a sudden demand occurs, such a scheme cannot effectively identify the real ability, nor can it predict potential risks through data fusion.

[0007] To solve the above technical problems, the present application provides the following technical solutions:

[0008] In a first aspect, the embodiments of the present application provide a multi-platform recruitment information management system based on big data, comprising a data acquisition module, a data processing module and a post matching module, further comprising:

[0009] A multi-source data fusion engine is configured to integrate structured resume data from a recruitment platform, operation log data from a production management system and real-time verification data from a third-party skill certification platform.

[0010] A dynamic capability assessment model is configured to extract potential skill features of a candidate based on an unstructured data parser, wherein the potential skill features include device operation trajectory pattern recognition results and emergency disposal scene simulation scores.

[0011] An elastic team formation strategy generator is configured to automatically generate a multi-dimensional skill combination scheme based on a process decomposition graph of an order production line, wherein the scheme includes at least matching parameters of core certified posts, trainable posts and experience transfer posts.

[0012] As a preferred scheme of the multi-platform recruitment information management system based on big data, the multi-source data fusion engine comprises:

[0013] A cross-platform certificate verification unit is configured to access an HACCP certification database and perform validity verification, and to synchronize and compare operation examination records of a certificate holder in the past 6 months during the verification.

[0014] A production behavior conversion unit is configured to convert Internet of Things sensor data of a food processing device into skill assessment indicators, including sorting speed standard deviation and device calibration response time.

[0015] An implicit skill mining unit is configured to analyze work-related video content in a three-party skill certification platform of a candidate and extract a tool use proficiency feature vector.

[0016] As a preferred scheme of the multi-platform recruitment information management system based on big data, in the production behavior conversion unit, the step of converting the Internet of Things sensor data of the food processing device into the skill assessment indicators comprises:

[0017] An average value of the sorting speed is calculated according to the following formula:

[0018]

[0019] wherein μ represents the average value of the sorting speed, N represents the total number of data acquisition, i.e., the number of continuously acquired sorting speed data, i represents a data index, and the value range is 1 to N, xi represents the i-th acquired sorting speed data. i

[0020] A sorting speed standard deviation indicator is constructed based on the average value according to the following formula:​

[0021]

[0022] wherein, sigma represents the sorting speed standard deviation, N represents the total number of data collection, i represents the data index, x i represents the i-th collected sorting speed data, mu represents the average value of the sorting speed;

[0023] For the device calibration response time, the response duration is calculated by the following formula for each calibration event:

[0024] wherein, represents the i1-th calibration response time, represents the i1-th calibration event completion time, represents the i1-th calibration event start time;

[0025] The response duration of all calibration events is integrated to obtain the average calibration response time, and the formula is:

[0026]

[0027] wherein, T c represents the average calibration response time, M represents the total number of calibration events, i1 represents the event index, represents the i1-th calibration response time.

[0028] As a preferred scheme of the multi-platform recruitment information management system based on big data, the dynamic capability evaluation model comprises:

[0029] The detachable certification adaptation module splits the complete qualification certification into N independent verification sub-skill units, N≥3, and each sub-skill unit corresponds to a specific process node on the production line.

[0030] The experience transfer degree calculation module constructs a cross-post skill conversion matrix based on historical employment data to calculate the probability value of non-certified personnel reaching the post requirements through targeted training.

[0031] The real-time attenuation warning module dynamically updates the skill effective coefficient by analyzing the latest three simulation operation data of the certified person.

[0032] As a preferred scheme of the multi-platform recruitment information management system based on big data, the detachable certification adaptation module performs the following operations:

[0033] a) Establish an authentication element mapping table to correspond the clauses of HACCP certification to specific operation nodes of the production line;

[0034] b) setting alternative ability proof rules for each operating node, including:

[0035] Unlicensed but with the same equipment preset hours of operation record more than the equivalent qualifications;

[0036] AR simulation test to achieve a set of level score can replace the written certificate requirements.

[0037] As a preferred scheme of the multi-platform recruitment information management system based on big data, wherein: in the detachable authentication adaptation module, the method of replacing the written certificate requirement by achieving a set of level score through AR simulation test includes score calculation, normalization processing and threshold judgment;

[0038] AR test adopts multi-item skill evaluation, and gives original score for each skill dimension. It is assumed that AR test contains k items, and the original scores of each item are respectively denoted as s j ,j=1,2,…,k, wherein, s j represents the original score of the jth AR test item, and k represents the number of AR test items;

[0039] On this basis, the weighted sum method is used to calculate the comprehensive score, and the formula is:

[0040]

[0041] wherein, s raw represents the comprehensive AR simulation original test score, w j represents the weight coefficient of the jth test item, reflecting the contribution of the item to the overall score;

[0042] Since the value range of each item score may be different, a linear normalization function is introduced to standardize each score, and the normalization formula is:

[0043] wherein, s ' j represents the normalized score of the jth test item, represents the minimum possible score of the jth test item, represents the maximum possible score of the jth test item;

[0044] The normalized score is substituted into the comprehensive score formula to form the normalized comprehensive score s:

[0045] wherein s represents the normalized comprehensive score;

[0046] Set threshold γ as the authentication standard, when the comprehensive normalized score satisfies s≥γ, it is considered to pass the authentication test, wherein γ represents the set level score threshold, which is used to judge whether the authentication meets the standard.

[0047] As a preferred scheme of the multi-platform recruitment information management system based on big data, the elastic team formation strategy generator comprises:

[0048] A process capability requirement parser automatically identifies key control points according to an order product specification and generates a skill requirement heat map;

[0049] A mixed team formation optimization algorithm matches the plasticity skill improvement direction for each non-certified personnel under the premise of meeting the core post certificate rate requirement;

[0050] A dynamic cost constraint module adjusts the combination ratio of personnel with different skill levels in real time in combination with the raw material price fluctuation curve and the order delivery time limit.

[0051] As a preferred scheme of the multi-platform recruitment information management system based on big data, in the dynamic cost constraint module, the unit cost of personnel at each level, the raw material price fluctuation, and the urgency of the order delivery time limit are introduced, the Softmax function is used to dynamically allocate the proportion of personnel at each skill level, and the combination ratio is defined as:

[0052]

[0053] Wherein, p o represents the combination ratio of personnel at skill level o, r represents the total number of skill levels, λ represents the sensitivity control factor of ratio adjustment, c o represents the unit cost of personnel at skill level p, α represents the weight coefficient of the influence of raw material price fluctuation on cost, C m represents the raw material price fluctuation coefficient, β represents the weight coefficient of the influence of order urgency on cost pressure, T d represents the remaining delivery time limit of the order, and the smaller the time limit value, the more urgent the order delivery.

[0054] In a second aspect, the present application provides a multi-platform recruitment information management method based on big data, comprising,

[0055] Step S1, constructing an industry feature enhancement database, integrating historical production anomaly records, equipment operation manuals, and industry safety specification texts of the target enterprise;

[0056] Step S2, deploying a dynamic skill radar scan, capturing actual operation pictures of candidates through a video analysis interface, and extracting tool holding angle stability and abnormal recognition reaction time indicators;

[0057] Step S3, generating an adaptive training path, outputting a shortest cycle standard training scheme according to the process bottleneck point of the current production line and the skill gap atlas of the candidate;

[0058] Step S4, implement a hierarchical access mechanism to divide the production line into core control area, auxiliary operation area and trainable area, and dynamically allocate work areas for personnel with different skill levels.

[0059] As a preferred scheme of the multi-platform recruitment information management method based on big data, in the step S3, the adaptive training path is generated by:

[0060] A skill unit correlation map is established to define the prerequisite relationship and the shortest training interval between sub-skills.

[0061] An augmented reality training scene library is developed, including a device failure simulation module and a sudden situation stress test module.

[0062] A capacity loss prediction model is deployed to seek an optimal balance point between training intensity and production progress.

[0063] The present application has the following advantages: the present application effectively reflects the production operation state through sorting speed, calibration response and other indicators, and uses AR simulation test to replace written certificates, realizes more intuitive and operable talent qualification certification, and makes up for the shortcomings of information isolation, subjective evaluation and untimely verification in traditional certification; the dynamic capability evaluation model and the elastic team formation strategy generator work together to match the post skills according to the actual production requirements, dynamically adjust the personnel combination ratio, and consider the cost and production progress, so as to effectively solve the problems of recruitment information fragmentation, data lag and skill evaluation deviation in the background, and improve the scientificity and accuracy of recruitment and training decision. BRIEF DESCRIPTION OF DRAWINGS

[0064] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0065] Figure 1 The frame schematic diagram of the multi-platform recruitment information management system in embodiment 1.

[0066] Figure 2 The flowchart of the multi-platform recruitment information management method in embodiment 1. DETAILED DESCRIPTION

[0067] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification.

[0068] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, it will be appreciated that the present application can be practiced in a variety of ways beyond the specific details set forth herein, having regard to the contents of the whole patent document, and that the present application can be practiced in other but essentially similar ways. Accordingly, the present application is not limited in scope to the specific implementations disclosed herein.

[0069] Secondly, the "one embodiment" or "embodiment" referred to herein means a specific feature, structure or characteristic that can be included in at least one implementation of the present application. "In one embodiment" appearing in different places in the specification does not mean the same embodiment, nor is it an embodiment that is independent of or mutually exclusive with other embodiments.

[0070] Embodiment 1, Reference Figure 1 and Figure 2 The embodiment provides a multi-platform recruitment information management system based on big data, which comprises a data acquisition module, a data processing module and a post matching module, and further comprises:

[0071] A multi-source data fusion engine is configured to integrate structured resume data from a recruitment platform, operation log data from a production management system and real-time verification data from a third-party skill certification platform.

[0072] The multi-source data fusion engine comprises:

[0073] A cross-platform certificate verification unit is configured to access an HACCP certification database and perform validity verification, and to synchronize and compare operation examination records of a certificate holder in the past 6 months during the verification.

[0074] A production behavior conversion unit is configured to convert Internet of Things sensor data of a food processing device into skill evaluation indexes, including a sorting speed standard deviation and a device calibration response time.

[0075] An implicit skill mining unit is configured to analyze work-related video content in a three-party skill certification platform of a candidate and extract a tool use proficiency feature vector.

[0076] In the production behavior conversion unit, the step of converting the Internet of Things sensor data of the food processing device into the skill evaluation indexes comprises:

[0077] The average value of the sorting speed is calculated according to the following formula:

[0078]

[0079] Wherein, μ represents the average value of the sorting speed, N represents the total number of data acquisition, i.e., the number of continuously acquired sorting speed data, i represents the data index, and the value range is 1 to N, xi represents the i-th acquired sorting speed data. i

[0080] ​The sorting speed standard deviation index is constructed based on the average value, and the formula is:

[0081]

[0082] Wherein, σ represents the sorting speed standard deviation, N represents the total number of data collection, i represents the data index, x i represents the i-th collected sorting speed data, μ represents the average value of the sorting speed;

[0083] For the device calibration response time, the response duration is calculated by the following formula for each calibration event:

[0084] Wherein, represents the i1-th calibration response time, represents the i1-th calibration event completion time, represents the i1-th calibration event start time;

[0085] The response duration of all calibration events is integrated to obtain the average calibration response time, and the formula is:

[0086]

[0087] Wherein, T c represents the average calibration response time, M represents the total number of calibration events, i1 represents the event index, represents the i1-th calibration response time;

[0088] Specifically, by processing the food processing equipment Internet of Things sensor data, the original data is converted into skill evaluation indexes describing the operation behavior, and the data processing method uses the average value and standard deviation calculation in statistics to reflect the device running stability according to the dynamic information collected by the sensor, and at the same time, the device response capability is revealed by the time interception calculation of the calibration event; The production operation behavior is expressed by quantitative indexes;

[0089] The dynamic capability evaluation model extracts the potential skill features of the candidate based on the unstructured data parser, and the potential skill features include the device operation trajectory pattern recognition result and the emergency disposal scene simulation score;

[0090] The dynamic capability evaluation model includes:

[0091] The detachable certification adaptation module splits the complete qualification certification into N independent verification sub-skill units, N≥3, and each sub-skill unit corresponds to a specific process node on the production line;

[0092] The experience transfer degree calculation module constructs a cross-post skill conversion matrix based on historical labor data to calculate the probability value of non-certified personnel reaching the post requirements through targeted training;

[0093] Real-time attenuation warning module, by analyzing the data of the last three times of simulation operation of the certificate holder, dynamically updating the skill effective coefficient;

[0094] The detachable authentication adaptation module performs the following operations:

[0095] a) Establish an authentication element mapping table to correspond the HACCP certification clauses to the specific operation nodes of the production line;

[0096] b) Set alternative ability proof rules for each operation node, including:

[0097] Those without a certificate but with more than preset hours of operation records on similar equipment are considered as equivalent qualifications;

[0098] Achieving a set level score through AR simulation test can replace the requirement for written certificates;

[0099] In the detachable authentication adaptation module, the method of replacing the requirement for written certificates through AR simulation test to achieve a set level score includes score calculation, normalization processing and threshold judgment;

[0100] AR test uses multiple skill evaluation, and gives an original score for each skill dimension. Assuming that AR test contains k items, the original scores of each item are denoted as s j ,j=1,2,…,k, where s j represents the original score of the jth AR test item, and k represents the number of AR test items;

[0101] On this basis, the weighted sum method is used to calculate the comprehensive score, and the formula is:

[0102]

[0103] where s raw represents the comprehensive AR simulation original test score, w j represents the weight coefficient of the jth test item, reflecting the contribution of the item to the overall score;

[0104] Since the value range of each item score may be different, a linear normalization function is introduced to standardize each score, and the normalization formula is:

[0105] where s ' j represents the normalized score of the jth test item, represents the minimum possible score of the jth test item, represents the maximum possible score of the jth test item;

[0106] The normalized score is substituted into the comprehensive score formula to form a normalized comprehensive score s:

[0107] where s represents the normalized comprehensive score;

[0108] A threshold value γ is set as the authentication standard, and when the comprehensive normalized score satisfies s≥γ, it is considered to pass the authentication test, where γ represents a set level score threshold value for judging whether the authentication meets the standard;

[0109] Specifically, the original scores of multiple AR test items are integrated into a comprehensive score by weighted summation, reflecting the importance of each skill dimension in overall ability evaluation. After introducing normalization processing, the influence of different test items due to the difference in value range is effectively alleviated. A set threshold value is used to judge whether the comprehensive score meets the authentication requirements, providing an intuitive reflection of the actual operation performance of the operator;

[0110] The elastic team strategy generator automatically generates a multi-dimensional skill combination scheme according to the process decomposition map of the order production line, which at least contains the matching parameters of core authentication positions, trainable positions and experience transfer positions;

[0111] The elastic team strategy generator includes:

[0112] The process capability requirement parser automatically identifies key control points according to the order product specification book and generates a skill demand heat map;

[0113] The mixed team optimization algorithm matches the plasticity skill improvement direction for each non-certified personnel under the premise of meeting the core position certificate rate requirement;

[0114] The dynamic cost constraint module combines the raw material price fluctuation curve and the order delivery time limit to adjust the combination ratio of personnel with different skill levels in real time;

[0115] In the dynamic cost constraint module, the unit cost of each level of personnel, the raw material price fluctuation and the urgency of the order delivery time limit are introduced, and the Softmax function is used to dynamically allocate the proportion of each skill level personnel. The combination ratio is defined as:

[0116]

[0117] where p o represents the combination ratio of skill level o personnel, r represents the total number of skill levels, λ represents the sensitivity control factor of ratio adjustment, c o represents the unit cost of skill level o personnel, α represents the weight coefficient of the influence of raw material price fluctuation on cost, C m represents the raw material price fluctuation coefficient, β represents the weight coefficient of the influence of order urgency on cost pressure, Td represents the remaining delivery time limit of the order, and the smaller the time limit value is, the more urgent the delivery of the order is;

[0118] Specifically, the cost factor and the order time limit are introduced into the dynamic adjustment of the personnel combination ratio, a soft maximum function is used to realize continuous adjustment of the ratio, the differences in unit cost of personnel are comprehensively considered, and under the combined action of fluctuations in the price of raw materials and the urgency of order delivery, the combination ratio of personnel with higher cost or order urgency will be relatively reduced, and the ratio of personnel with competitive advantage will be correspondingly increased. Through real-time updating of relevant parameters, the model can adapt to market dynamic changes, so that the balance between production cost and delivery time of the system is optimized.

[0119] The embodiment also provides a management method of the above-mentioned multi-platform recruitment information management system based on big data, comprising the following steps: S1, constructing an industry feature enhancement database, integrating historical production anomaly records, device operation manuals and industry safety specification texts of a target enterprise;

[0120] S2, deploying a dynamic skill radar scan, capturing actual operation pictures of candidates through a video analysis interface, and extracting tool holding angle stability and abnormal identification reaction time indicators of the candidates;

[0121] S3, generating an adaptive training path, outputting a shortest-period training scheme according to a process bottleneck point of a current production line and a skill gap atlas of a candidate;

[0122] S4, implementing a hierarchical access mechanism, dividing the production line into a core control area, an auxiliary operation area and a trainable area, and dynamically allocating work areas of personnel with different skill levels;

[0123] The generation of the adaptive training path in step S3 comprises the following steps:

[0124] S301, establishing a skill unit correlation atlas, and defining prerequisite relationships and shortest training intervals between sub-skills;

[0125] S302, developing an augmented reality training scene library, comprising a device fault simulation module and a sudden situation stress test module;

[0126] S303, deploying a production capacity loss prediction model, and seeking an optimal balance point between training intensity and production progress;

[0127] In summary, the skill certification system, the dynamic capability migration evaluation model and the flexible production team strategy constructed by the present application effectively solve the impossible triangle of quality, cost and time in the seasonal order scenario. Unlike the rigid screening logic of traditional systems, the present application realizes fine management of skill elements and dynamic adaptation of production resources, and is particularly suitable for fields such as fresh food processing and export food, which have strict requirements on operation specification and labor flexibility.

[0128] It should be noted that the above examples are only used to illustrate the technical solutions of the present application but not to limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced without departing from the spirit and scope of the present application, and all modifications and equivalents should be included in the scope of the claims of the present application.

Claims

1. A big data-based multi-platform recruitment information management system comprising a data collection module, a data processing module and a post matching module, characterized in that, Also comprising: a multi-source data fusion engine for integrating structured resume data from recruitment platforms, operation log data from production management systems, and real-time verification data from third-party skill certification platforms; a dynamic capability assessment model based on an unstructured data parser to extract potential skill features of candidates, including device operation trajectory pattern recognition results and emergency disposal scenario simulation scores; an elastic team formation strategy generator to automatically generate multi-dimensional skill combination schemes based on the process decomposition map of the order production line, including at least the ratio parameters of core certified positions, trainable positions, and experience transfer positions; the elastic team formation strategy generator comprises: a process capability demand parser to automatically identify key control points based on the order product specification and generate a skill demand heat map; a hybrid team optimization algorithm to match the plasticity skill improvement direction for each non-certified person under the premise of meeting the core position certificate rate requirement; a dynamic cost constraint module to adjust the combination ratio of personnel with different skill levels in real time in combination with the raw material price fluctuation curve and the order delivery time limit.

2. The big data based multi-platform recruitment information management system as claimed in claim 1, wherein: The multi-source data fusion engine comprises: a cross-platform certificate verification unit that accesses the HACCP certification database and performs validity verification, and synchronously compares the operation examination records of the certificate holder in the past 6 months during verification; a production behavior conversion unit that converts Internet of Things sensor data of food processing equipment into skill evaluation indicators, including sorting speed standard deviation and equipment calibration response time; an implicit skill mining unit that analyzes work-related video content in the candidate's three-party skill certification platform and extracts a tool use proficiency feature vector.

3. The big data based multi-platform recruitment information management system as claimed in claim 2, wherein: In the production behavior conversion unit, the step of converting the Internet of Things sensor data of food processing equipment into skill evaluation indicators comprises: calculating the average value of the sorting speed, the formula is: ， wherein, represents the average value of the sorting speed, represents the total number of data collection, i.e. the number of continuously collected sorting speed data, represents the data index, the value range is 1 to , represents the sorting speed data collected at the th constructing the sorting speed standard deviation index based on the average value, the formula is: ， in, This represents the standard deviation of sorting speed. Indicates the total number of data collected. Indicates a data index. Indicates the first The collected sorting speed data, This represents the average sorting speed. For the equipment calibration response time, the response duration is calculated for each calibration event by the following formula: wherein, denotes the time of the calibration event, denotes the time of the start of the calibration event; Calculate the average calibration response time by summing up the response durations of all calibration events, the formula is: ， wherein, represents the average calibration response time, represents the total number of calibration events, represents the event index, represents the first calibration response time.

4. The big data based multi-platform recruitment information management system as claimed in claim 1, wherein: The dynamic capability assessment model comprises: a detachable certification adaptation module that splits the complete qualification certification into N independent verification sub-skill units, N≥3, each sub-skill unit corresponding to a specific process node on the production line; an experience transfer degree calculation module that constructs a cross-post skill conversion matrix based on historical employment data to calculate the probability value of non-certified personnel reaching the job requirements through targeted training; a real-time decay warning module that dynamically updates the skill validity coefficient of the certificate holder by analyzing their last 3 simulation operation data.

5. The big data based multi-platform recruitment information management system as claimed in claim 4, wherein: The detachable certification adaptation module performs the following operations: a) Establish an authentication element mapping relationship table to correspond the clauses of HACCP certification to specific operation nodes on the production line; b) Set alternative capability proof rules for each operation node, including: Non-certified but with more than a preset number of hours of operation records on similar equipment are considered equivalent qualifications; Achieving a set level score through AR simulation testing can replace the requirement for a written certificate.

6. The big data based multi-platform recruitment information management system as claimed in claim 5, wherein: The disassemblable authentication adaptation module, through AR simulation test to reach the set level score can replace the written certificate requirement mode includes: score calculation, normalization processing and threshold judgment; The AR test employs multiple skill assessments, providing a raw score for each skill dimension. Let the AR test include... Each of the following items has a raw score recorded as follows: ,in, Indicates the first The original scores of each AR test item Indicates the number of AR test items; On this basis, the weighted sum method is used to calculate the comprehensive score, and the formula is: ， wherein, represents the overall AR simulation raw test score, represents the weight coefficient of the test item, reflecting the contribution of the item to the overall score; Due to the difference in the value range of each item score, a linear normalization function is introduced to standardize each item score, and the normalization formula is: wherein, represents the normalized score of the th test item, represents the minimum possible score of the th test item, represents the maximum possible score of the th test item; The normalized scores are substituted into the comprehensive score formula to form a normalized comprehensive score : wherein denotes the normalized overall score; Setting threshold As a standard for authentication, when the comprehensive normalized score satisfies , it is considered to pass the authentication test, wherein represents a set level score threshold for determining whether the authentication is up to standard.

7. The big data based multi-platform recruitment information management system as claimed in claim 1, wherein: In the dynamic cost constraint module, the unit cost of each level of personnel, the price fluctuation of raw materials and the urgency of order delivery time limit are introduced, the Softmax function is used to dynamically allocate the proportion of each skill level personnel, and the combination ratio is defined as: ， wherein, represents the skill level represents the combination ratio of personnel, represents the total number of skill levels, represents the sensitivity control factor of the ratio adjustment, represents the skill level represents the unit cost of personnel, represents the weight coefficient of the influence of raw material price fluctuation on cost, represents the raw material price fluctuation coefficient, represents the weight coefficient of the influence of order urgency on cost pressure, represents the order remaining delivery time limit, the smaller the time limit value, the more urgent the order delivery.

8. A big data-based multi-platform recruitment information management method based on the big data-based multi-platform recruitment information management system of any one of claims 1-7, characterized in that, Including: Step S1, construct an industry feature enhancement database, integrate the target enterprise's historical production anomaly records, equipment operation manual and industry safety specification text; Step S2, deploy dynamic skill radar scanning, capture the candidate's actual operation picture through video analysis interface, extract the tool holding angle stability, abnormal identification reaction time index; Step S3, generate adaptive training path, according to the process bottleneck point of the current production line and the skill gap atlas of the candidate, output the shortest period training scheme; Step S4, implement the hierarchical access mechanism, divide the production line into core control area, auxiliary operation area and trainable area, dynamically allocate the work area of personnel with different skill levels.

9. The big data-based multi-platform recruitment information management method of claim 8, wherein: The generation of adaptive training path in step S3 includes: Establish the skill unit association map, define the prerequisite relationship and the shortest training interval between each sub skill; Develop augmented reality training scene library, including equipment failure simulation module and emergency situation stress test module; Deploy the production capacity loss prediction model to find the optimal balance point between training intensity and production progress.

Images