A cloud platform-based intelligent labor union management system

Abstract

The present application relates to the field of trade union management, and discloses a smart trade union management system based on a cloud platform, which is used to improve the fine level of trade union management of industrial enterprises. A plurality of independent trade union management scenarios are defined on an industrial cloud platform, a scenario-based feature function is defined for each scenario, the required behavior data types are determined, and a mapping relationship is established. Then, trade union member behavior data is continuously collected and processed into standardized data. For each trade union member in each scenario, a dynamic contribution index is obtained by using the cooperative game theory and the Monte Carlo simulation method in combination with the standardized data and the scenario-based feature function, and a comprehensive contribution portrait of the trade union member is formed after aggregation. According to the portrait, the industrial cloud platform automatically completes management operations such as welfare distribution. The present application ensures the scientificity and dynamic adaptability of management decisions, and improves the trade union management level and the member satisfaction.

Description

Technical Field

[0001] This invention relates to the field of trade union management, and in particular to a smart trade union management system based on a cloud platform. Background Technology

[0002] As industrial enterprises continue to expand and their workforce becomes increasingly diversified, the complexity and importance of trade unions, as a crucial bridge between enterprises and employees, are becoming increasingly prominent. Traditional trade union management models, relying primarily on manual statistics and experience-based judgment, have gradually revealed inefficiencies and inaccuracies in handling large amounts of member information, organizing various activities, and allocating resources. For example, in the allocation of welfare resources, it is often difficult to comprehensively and objectively consider the actual contributions of each member, resulting in allocation outcomes that fail to satisfy members universally, thereby affecting members' sense of identification with the trade union and their enthusiasm for participation.

[0003] Driven by the digital wave, emerging technologies such as cloud computing and big data have developed rapidly, especially the widespread application of industrial cloud platforms, bringing new opportunities and challenges to trade union management. Leveraging the powerful computing capabilities, underlying device connectivity, and massive industrial data storage capacity of industrial cloud platforms, trade unions can efficiently process and analyze massive amounts of behavioral data from their members (especially production line workers and technical personnel). Therefore, exploring a smart trade union management system based on industrial cloud platforms to achieve interconnectivity between production data and trade union management data has become an inevitable choice to adapt to the development trend of the digital age and improve the level of trade union management.

[0004] Traditional union management methods typically employ uniform standards and models, failing to adequately consider the differences and characteristics of member behavior in various management scenarios. In activity organization and welfare distribution scenarios, members' contribution methods and evaluation criteria may differ drastically, but traditional methods often fail to distinguish these differences, resulting in poor management effectiveness and an inability to meet the diverse needs of members in different scenarios. Traditional trade union management often relies on subjective evaluations or simple statistical indicators when assessing members' contributions. It lacks scientific and objective evaluation methods, is easily influenced by human factors, and cannot accurately reflect members' actual contributions. This leads to problems such as unfair distribution of welfare resources and unreasonable evaluation results, which affect members' enthusiasm and the cohesion of the trade union organization. The existing trade union management model lacks a dynamic adjustment mechanism in resource allocation and management decision-making, making it difficult to adjust management strategies in a timely manner according to changes in member behavior and actual needs. In the process of welfare resource allocation, once the allocation plan is determined, it is often difficult to make flexible adjustments based on the subsequent performance of members, resulting in a disconnect between resource allocation and members' actual contributions, and failing to give full play to the incentive role of resources. Therefore, we propose a cloud-based smart trade union management system to address the aforementioned issues. Summary of the Invention

[0005] This invention provides a cloud-based intelligent trade union management system to improve the level of precision in trade union management.

[0006] The first aspect of this invention provides a cloud-based smart trade union management system, comprising: a scenario definition module for defining multiple independent trade union management scenarios in the cloud platform and defining a scenario-based feature function for each management scenario; a data processing module for continuously collecting behavioral data of all trade union members and processing the behavioral data into standardized data; a contribution index module for processing the standardized data using the corresponding scenario-based feature function for each trade union member in each management scenario to obtain the dynamic contribution index of the trade union member in the current scenario; a comprehensive profile module for aggregating the dynamic contribution indices of each trade union member in all management scenarios to form a comprehensive contribution profile of the trade union member; and a management operation module for driving the cloud platform to complete at least one management operation based on the comprehensive contribution profile of the trade union member, including differentiated allocation of welfare resources, targeted scheduling of activity resources, and ranking and granting of awards.

[0007] Optionally, in a first implementation of the first aspect of the present invention, the method includes: determining multiple independent trade union management business dimensions to form multiple independent management scenarios corresponding to different management objectives; defining a scenario-based feature function for each management scenario, wherein all scenario-based feature functions constitute a function set, wherein each function is configured to receive data from a subset of trade union members and output a quantitative value representing the overall performance of the subset in the scenario; configuring the specific behavioral data type required for the calculation of each scenario-based feature function according to the evaluation objectives of each management scenario, and establishing a mapping relationship between the behavioral data type and the scenario-based feature function.

[0008] Optionally, in a second implementation of the first aspect of the present invention, the method includes: collecting original behavioral records generated by all union members in various activities and interactions to form an original behavioral data set; based on the established mapping relationship, selecting the behavioral data necessary for the calculation of the corresponding scenario-based feature function from the original behavioral data set for each management scenario to obtain a dataset to be processed for each scenario; processing the behavioral data in each dataset to be processed to obtain a standardized data set for each scenario; and processing the numerical data in each standardized data set to obtain standardized data that can be directly used by each scenario-based feature function.

[0009] Optionally, in the third implementation of the first aspect of the present invention, participation records of union members participating in various union activities, formal proposals and suggestions submitted by union members, and initiation and completion records of mutual assistance behaviors among union members are extracted from the business database of the cloud platform to obtain categorized behavior records; each record in the categorized behavior records is marked with its occurrence timestamp, and quantifiable evaluation indicators are identified to form quantified behavior data with time-series markers; based on the quantified behavior data with time-series markers, periodic statistical values ​​for each union member in each type of behavior are calculated for each member in a preset statistical period to obtain union member periodic behavior baseline data; according to the target time window corresponding to the current calculation of the contribution index, statistical values ​​of all relevant periods covering the time window are aggregated from the union member periodic behavior baseline data to form an original behavior data set.

[0010] Optionally, in the fourth implementation of the first aspect of the present invention, the method includes: calculating the contribution index of the current union member in the current management scenario, setting the number of iterations of the Monte Carlo simulation to form simulation configuration parameters; obtaining a random permutation of the complete set of union members for a specified number of times based on the simulation configuration parameters, forming a random permutation sequence set; for each permutation in the random permutation sequence set, calculating the change in scenario performance value brought about by the current union member joining the subset of union members in front of it based on the standardized data and the corresponding scenario-based feature function, obtaining the marginal contribution value of a single simulation, and forming a marginal contribution set of all simulated marginal contribution values; calculating the dynamic contribution index of the union member in the current management scenario by calculating all marginal contribution values ​​in the marginal contribution set.

[0011] Optionally, in the fifth implementation of the first aspect of the present invention, based on the standardized data, the basic performance ranking of all union members in the current management scenario is calculated to obtain an initial ranking sequence of union members; when obtaining each random arrangement, the positions of adjacent union members in the initial ranking sequence are swapped according to a preset perturbation probability to form a basic perturbation arrangement; a global shuffle operation is applied to the basic perturbation arrangement to ensure that all union member positions have the opportunity to be redistributed to obtain a final single arrangement that meets the randomness requirement; the process of obtaining the basic perturbation arrangement and applying the global shuffle operation is repeated until the set of random arrangement sequences is obtained, wherein each final single arrangement obtained is used as the union member order basis for the marginal contribution calculation.

[0012] Optionally, in a sixth implementation of the first aspect of the present invention, the preset perturbation probability is obtained by the following adaptive calculation formula: ; in, This represents the probability that two adjacent union members will swap positions in the current simulation. The system's preset basic disturbance constant; , V represents the base performance value of two adjacent players; avg This represents the average basic performance of the union members in this group.

[0013] Optionally, in the seventh implementation of the first aspect of the present invention, the automatic completion of management operations by the driving cloud platform includes: extracting corresponding contribution data from the comprehensive contribution profile of the union members according to the type of management operation to be executed, and obtaining management decision input data; inputting the management decision input data into a preset business decision rule set to obtain a set of management parameters including specific allocation quotas, scheduling priorities or sorting results; and obtaining a set of executable instructions for the corresponding business subsystem in the driving cloud platform to perform specific operations based on the set of management parameters.

[0014] Optionally, in the eighth implementation of the first aspect of the present invention, an operation feedback module is further included: collecting feedback data from union members on the management operation, and associating the feedback data with the management parameter set of the corresponding operation to obtain an operation feedback record; based on the operation feedback record, analyzing the correlation between the union member contribution index and feedback satisfaction under different management scenarios to obtain scenario weight adjustment suggestion data; providing a configuration interface in the cloud platform to receive the administrator's confirmation or modification of the scenario weight adjustment suggestion data, and updating the scenario weight configuration parameters accordingly.

[0015] Optionally, in the ninth implementation of the first aspect of the present invention, the analysis of the correlation between the union member contribution index and feedback satisfaction under different management scenarios includes: for each management operation, calculating the deviation between the actual management result and the union member's expectations based on the union member's satisfaction feedback in the operation feedback record, and obtaining satisfaction deviation data; performing correlation analysis between the satisfaction deviation data and the corresponding dynamic contribution index used when performing the operation, calculating the utility coefficient of the contribution index, and forming a correlation analysis result; based on the correlation analysis result, evaluating the effectiveness of the dynamic contribution index of each management scenario under the current decision-making rules, calculating the adjustment change value required for the weight of each scenario; and comprehensively obtaining scenario weight adjustment suggestion data to guide the correction of the relative importance of each scenario by combining the evaluation results of all management operations.

[0016] The mechanism of this invention is as follows: It breaks through the subjectivity and static limitations of the traditional fixed-weight scoring method, realizes a fair, interpretable and dynamically adaptive quantitative assessment of the value of union members' behavior, and significantly improves the level of refinement of union management, the scientific nature of resource allocation and the dynamic adaptability of the system. Beneficial effects: By defining multiple independent union management scenarios, diverse participation channels are provided for union members. Behavioral data of union members in different scenarios is accurately collected and analyzed to obtain a dynamic contribution index and comprehensive contribution profile. This allows union members to clearly understand their role and contribution within the union, enabling them to participate in union activities more effectively, fostering positive interaction with the union system, enhancing their sense of identity and belonging to the union, and improving overall participation and collaboration. Based on the comprehensive contribution profile of union members, the union can accurately understand the needs and contribution levels of union members, and achieve a reasonable allocation of welfare resources, activity resources, etc. This resource allocation method matches the actual contributions and needs of union members, avoids waste and unreasonable allocation of resources, enables union members and the union to achieve efficient collaboration in resource utilization, and improves the efficiency of resource use. With the support of the industrial cloud platform, union member behavior data collected from different management scenarios is deeply shared with industrial business data, providing rich material for obtaining dynamic contribution indices and comprehensive contribution profiles, making the evaluation results more comprehensive and accurate. At the same time, it breaks down data silos between production management and union management, improving the overall efficiency and scientific nature of union management. Smart union management technology helps enhance the organizational power and influence of unions. Within the industrial sector, various union organizations can leverage smart union management technology based on the industrial cloud platform to achieve information sharing and cross-enterprise experience exchange. Smart trade union management technology helps enhance the organizational capacity and influence of trade unions. Within the industry, trade union organizations can leverage this technology to achieve information sharing and experience exchange. By analyzing the contribution data and management models of different trade union members, it promotes collaborative development across the entire industry, improves the overall level of the industry, and creates more favorable conditions for industry development. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of an embodiment of the cloud-based smart trade union management system of the present invention.

[0018] Figure 2 This is a schematic diagram of another embodiment of the cloud-based smart trade union management system of the present invention.

[0019] Figure 3 This is a schematic diagram of one embodiment of a smart trade union management device based on a cloud platform according to the present invention. Detailed Implementation

[0020] This invention provides a cloud-based smart trade union management system to improve the precision of trade union management. The terms "first," "second," "third," "fourth," etc. (if applicable) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0021] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the cloud-based smart trade union management system of the present invention includes: 101. The scenario definition module is used to define multiple independent union management scenarios in the cloud platform and define a scenario-based feature function for each management scenario. Each scenario-based feature function is used to calculate the quantitative performance value of any subset of union members under that management scenario.

[0022] It is understood that the executing entity of this invention can be a cloud-based smart trade union management device, a terminal, or a server; no specific limitation is made here. This embodiment of the invention will be described using a server as an example.

[0023] It should be noted that the system divides the management into three completely independent and non-interfering scenarios: technological innovation and mentorship, cultural and sports activities organization and participation, and democratic management and suggestions. For each of these three scenarios, the system defines scenario-based feature functions to evaluate the total quantitative performance value (absolute base score) created by any combination of employees (a subset of union members) in collaboration under a specific scenario.

[0024] In the context of technological innovation and mentorship, a feature function is used to evaluate the team's overall performance in solving technical challenges. Taking three core union members in a work group as an example: Senior Technician A, Junior Technician B, and Equipment Operator C. If working individually, the feature function assigns A 60 points for solving a routine problem, B 20 points, and C 10 points. If A and B team up (a two-person subset), a mentorship bonus is triggered, increasing the performance score to 95 points. If A, B, and C form a complete problem-solving team (a three-person subset), a large-scale cross-job problem-solving mechanism is triggered, and the total performance score for this subset is set at 130 points.

[0025] In the context of organizing and participating in cultural and sports activities, a feature function is used to evaluate the activity coverage and basic satisfaction scores. For example, organizing a basketball game: employee D, with athletic talent, scores 40 points if responsible alone, while employee E, with promotional talent, scores 30 points if responsible alone. If D and E form a team (a subset of two), the synergistic effect of professional organization and extensive promotion results in a performance value of 90 points for this subset assigned by the feature function.

[0026] In a scenario involving democratic management and suggestions for improvement, a characteristic function is used to evaluate the basic score of the proposal's effectiveness. Employee A and Employee D each propose energy-saving suggestions; A's individual proposal receives 30 points, and D receives 20 points. If A and D conduct a cross-departmental joint investigation and form a complete water cycle renovation proposal (a subset of their work), the characteristic function determines that its global relevance is extremely high, resulting in a performance value of 85 points.

[0027] 102. Data processing module, used to continuously collect behavioral data of all union members and process the behavioral data into standardized data that meets the requirements of characteristic function calculation for each scenario.

[0028] It should be noted that the cloud platform, as the data hub, captures unstructured daily business traces in real time, cleans and reduces their dimensions, and transforms them into standardized labels and parameters that can be read by the feature function in step 101.

[0029] Logs retrieved from the Production Execution System show that Senior Technician A, Junior Technician B, and Equipment Operator C jointly handled a workshop downtime fault, which took 16 hours. The platform's parsing engine processed this into standardized data: For Technician A, the data includes [Difficulty: Advanced], [Role: Leading], [Mentorship Collaboration Trigger: Yes], and [Work Hours Involved: 16]; for Technician B, the data includes [Role: Assisting] and [Mentorship Collaboration Trigger: Yes]; and for Technician C, the data includes [Role: External Support].

[0030] The cloud platform scraped basketball game data from the union's mini-program: Ding was responsible for registration (50 people checked in), and Wu was responsible for promotion (200 likes), with an average satisfaction rating of 4.8 stars. The platform normalized the data, converting Ding's behavior into [behavior category: activity initiation], [coverage scalar: 85], and [satisfaction weight: 1.2]; and Wu's behavior into [behavior category: activity promotion], [influence scalar: 92], and [satisfaction weight: 1.2].

[0031] The platform retrieves the "Workshop Water Circulation Renovation Plan" jointly submitted by A and D in the OA system, along with the subsequent financial surplus report (saving 35,000 yuan). The platform maps these to standard indicators, providing A and D with the following consistent results: [Proposal Category: Energy Conservation and Emission Reduction], [Collaboration Attribute: Cross-departmental], and [Benefit Conversion Index: 75].

[0032] 103. The contribution index module is used to obtain the dynamic contribution index of each union member in each management scenario. Based on cooperative game theory and Monte Carlo simulation, it processes standardized data using the corresponding scenario-specific feature function to obtain the dynamic contribution index of the union member in the current scenario.

[0033] It should be noted that the system uses Monte Carlo simulation, randomly sampling and queuing real team union members who actually participate in the task, calculating marginal contributions, and finally mapping them to a dynamic contribution index on a percentage scale.

[0034] Taking a senior technician, A, in a technological innovation scenario as an example: In handling a downtime malfunction, the actual participants are A, B, and C. The system needs to determine A's absolute contribution to the total score of 130 points. The cloud platform conducts multiple rounds of random Monte Carlo simulations to determine the order in which these three individuals join the task.

[0035] Simulation Sequence A: The system randomly generates the team formation order as C -> B -> A. C joins, contributing 10 points to the characteristic function; then B joins, and the combined performance of C and B is calculated as 35 points; finally, A joins (C + B + A), bringing the total team score to 130 points. In this simulation, A's addition increased the score from 35 to 130 points, resulting in a marginal contribution of 95 points.

[0036] Simulated sequence B: The order is A -> C -> B. A enters first and scores 60 points; C joins (A + C) and scores 75 points; B joins and scores 130 points. A's marginal contribution is 60 points.

[0037] After thousands of randomized simulations involving only three individuals (A, B, and C), the system determined that A's average marginal absolute contribution score was 75. The cloud platform then normalized this absolute score to a percentage based on the overall level of the department for the current quarter, calculating that senior technician A's dynamic contribution index in this scenario was 94.5 (out of 100).

[0038] 104. The comprehensive profile module is used to aggregate the dynamic contribution index of each union member under all management scenarios to form a comprehensive contribution profile of the union member.

[0039] It should be noted that the cloud platform performs multi-dimensional weighting and structural analysis on the indices of a single scenario to create a precise digital self-portrait for union members. The system presets the annual scenario weights as follows: technological innovation (50%), cultural and sports activities (20%), and democratic management (30%).

[0040] Retrieve the dynamic contribution indices of senior technician A and employee D, which have been converted into percentages, in three scenarios, and perform a weighted calculation: Data aggregation for Senior Technician A: Technical Index (94.5), Cultural and Sports Index (60.0, at a basic participation level), and Democratic Index (88.0). Weighted total score = (94.5 × 0.5) + (60.0 × 0.2) + (88.0 × 0.3) = 47.25 + 12.0 + 26.4 = 85.65 points.

[0041] The profile shows an overall score of 85.65, ranking in the top 5% of the entire factory. The system identified an extremely high technical dimension index (94.5), representing the highest peak for any individual. The core tags identified are: Technical Leader, Outstanding Core Member.

[0042] Employee Ding's data aggregation: Technical Index (65.0, at the basic standard level), Cultural and Sports Index (96.0), Democratic Index (82.0). Weighted total score = (65.0 × 0.5) + (96.0 × 0.2) + (82.0 × 0.3) = 32.5 + 19.2 + 24.6 = 76.3 points.

[0043] The profile yielded an overall score of 76.3, ranking in the top 15% of the entire factory. The system identified a cultural and sports index as high as 96.0, far exceeding the baseline. Key tags identified: Cultural and sports organization expert, active core member.

[0044] 105. Management Operation Module, used to drive the cloud platform to automatically complete at least one of the following management operations based on the comprehensive contribution profile of union members: differentiated allocation of welfare resources, targeted scheduling of activity resources, and ranking and granting of awards.

[0045] It should be noted that the cloud platform has a built-in strategy engine that automatically executes management actions based on the accurate profile and comprehensive score obtained in step 104, thereby achieving efficient closed-loop resource allocation.

[0046] In the quarterly Union Star selection, the system automatically retrieves the comprehensive score list. A ranked at the top with 85.65 points and also has the "Technology Leader" tag. The system directly fills in his / her first-priority candidate form for the quarterly Innovation Pioneer, eliminating the need for manual application and sending him / her directly to the union chairman's approval terminal.

[0047] The platform differentiates the allocation of welfare resources. It detected that Person A is in the top 5% of the entire factory's elite core group and automatically upgraded their annual routine physical examination privileges to a high-end health and wellness package. Simultaneously, the system detected Person D's "sports and cultural organization expert" tag and a high sports and cultural index of 96.0, automatically triggering a special reward: a brand-name badminton racket set was allocated from the welfare pool and sent to Person D's workstation via internal express delivery.

[0048] The system strategically allocated resources for the event. Based on A's technological leadership, A was automatically placed in a high-level seat at the next month's Industry 4.0 international online seminar. For D, the platform automatically created a VIP account for online training in large-scale event planning and public relations, based on D's artistic and athletic strengths, to facilitate their skill development.

[0049] The automation of management execution is compared in Table 1 below: Table 1

[0050] Please see Figure 2 Another embodiment of the cloud-based smart trade union management system of the present invention includes: 201. The scenario definition module is used to define multiple independent union management scenarios in the cloud platform and define a scenario-based feature function for each management scenario. Each scenario-based feature function is used to calculate the quantitative performance value of any subset of union members under that management scenario.

[0051] Specifically, multiple independent dimensions of trade union management business are identified, forming multiple independent management scenarios corresponding to different management objectives. A scenario-based feature function is defined for each management scenario, and all scenario-based feature functions constitute a function set. Each function is configured to receive data from a subset of trade union members and output a quantitative value representing the overall performance of that subset in that scenario. Based on the evaluation objectives of each management scenario, the specific behavioral data types required for the calculation of each scenario-based feature function are configured, and a mapping relationship between behavioral data types and scenario-based feature functions is established. The multiple independent management scenarios, function sets, and mapping relationships together serve as the configuration basis for subsequent data standardization and contribution index calculation.

[0052] It should be noted that in the early stages of building the smart union cloud platform for a large high-end manufacturing enterprise, the cloud platform architect and the union management jointly sorted out the union's complex daily core business. To avoid cross-contamination of evaluation dimensions, they extracted three independent, clearly defined, and non-interfering management scenarios: First, the skills enhancement and mentorship scenario, which focuses purely on the company's core business skills training, participation in technical competitions, and pairing of new and old employees for technical assistance. Second, the employee care and volunteer service scenario, which strips away the technical attributes and focuses specifically on activities that embody humanistic care, such as helping employees in difficulty, providing psychological counseling, and providing mutual assistance among volunteers in their spare time; Third, the scenario of democratic suggestions and management supervision, which is positioned to implement the rationalization suggestions made by employees to the management on process optimization, energy conservation and emission reduction, and the proposals of the employee representative assembly.

[0053] For each of these three scenarios, a strongly correlated behavioral data type was bound, forming an exclusive mapping dictionary. For the skills enhancement scenario, the system limits the reading to hard indicators such as internal training course duration, skills competition award level, and effective duration of mentorship agreements; for the employee care scenario, the system limits the reading to soft indicators such as mutual aid fund donation amount, visit and condolence visit to employees in need check-in time, and number of volunteer activities initiated; for the democratic suggestion scenario, the system strictly maps the number of proposals submitted, the proposal adoption level (team / department / company), and the direct economic benefit audit results.

[0054] Taking a skills enhancement and mentorship scenario as an example, the system evaluates senior engineer Zhang, equipment operator Sun, and newly hired assistant engineer Li. The underlying rules of the feature function are set as follows: if Zhang is a single-person subset, the output base quantization value is 50 based on his teaching data; the output for Sun's single-person subset is 30; and the output for Li's single-person subset is 20. However, when evaluating the two-person subset (Zhang + Li), if the feature function finds a valid mentor-apprentice pairing agreement in the mapping data, the system determines that a higher-order synergistic amplification effect of the mentor-apprentice pairing is triggered, and the overall quantization value of this subset jumps to 110 (far greater than 50 + 20).

[0055] Similarly, in the scenario of democratic suggestion, we set the quantification value of Zhou's individual proposal to 30, and the quantification value of Zhao's individual proposal to 30. If Zhou and Zhao form a subset, the feature function recognizes that this is a heavyweight proposal that is jointly signed across departments and adopted at the company level. It determines that it breaks down departmental barriers and generates global value, and outputs the dimensionless performance value of this joint subset as 95. Through this step, the cloud platform solidifies the complex business logic into a mathematical model.

[0056] 202. Data processing module, used to continuously collect behavioral data of all union members and process the behavioral data into standardized data that meets the requirements of characteristic function calculation for each scenario.

[0057] Specifically, the system collects raw behavioral records of all union members during various activities and interactions from the business database of the cloud platform, forming a raw behavioral data set. Based on the established mapping relationship, it selects the behavioral data necessary for calculating the corresponding scenario-based feature function for each management scenario from the raw behavioral data set, obtaining the dataset to be processed for each scenario. The behavioral data in each dataset to be processed is cleaned and formatted, invalid records are removed, and heterogeneous data is converted into a unified format, resulting in a standardized data set for each scenario. The numerical data in each standardized data set is normalized and converted into dimensionless values ​​within a predetermined range, resulting in standardized data that can be directly used by the scenario-based feature function.

[0058] Furthermore, from the cloud platform's business database, participation records of union members participating in various union activities, formal proposals and suggestions submitted by union members, and initiation and completion records of mutual assistance behaviors among union members are extracted to obtain categorized behavior records. Each record in the categorized behavior records is marked with its timestamp, and quantifiable evaluation indicators are identified to form time-series-marked quantitative behavior data. Based on the time-series-marked quantitative behavior data, periodic statistical values ​​for each union member in each type of behavior are calculated in units of a preset statistical period to obtain the union member periodic behavior baseline data. According to the target time window corresponding to the current contribution index calculation, statistical values ​​of all relevant periods covering the time window are aggregated from the union member periodic behavior baseline data to form the original behavior data set used for feature function calculation.

[0059] It should be noted that, to ensure the continuity of subsequent calculations, the current target time window is set as the first quarter of 2026 (January 1 to March 31).

[0060] The cloud platform continuously retrieves raw logs from the enterprise's Human Resource Management (HRM), Office Automation (OA), Access Control and Attendance System, and Trade Union WeChat Mini Program via API interfaces at a T+1 frequency. Based on the strict mapping relationship established by the 201 steps, the system accurately sorts the massive amount of data: Zhang's advanced CAD course in February and Sun's equipment maintenance training in March in the HR system are classified into the Skills Improvement pending processing set; Wang's four visits to nursing homes in the Mini Program are classified into the Employee Care pending processing set; and Zhou and Zhao's jointly submitted R&D optimization plan, which was adopted, is classified into the Democratic Suggestion pending processing set.

[0061] The platform performs rule-based verification on the data to be processed to ensure authenticity. In skills enhancement scenarios, if the system identifies a record of online training where the IP address is suspected of being logged in simultaneously from different locations, or the viewing time is less than 10% of the total time, it automatically marks the record as invalid and removes it. Subsequently, the system uniformly transforms the retained valid heterogeneous data (text logs, GPS check-ins, approval forms) into categorized behavior records with precise standard timestamps.

[0062] To completely eliminate the incomparability caused by physical dimensions (hours, times, yuan), the system performs a linear transformation on the cleaned values.

[0063] For skills enhancement scenarios: The system extracted that Senior Engineer Zhang conducted 3 high-quality lectures in the first quarter. The system converted the data based on a dimensionless benchmark of 60 points per lecture, resulting in a baseline data of 180.0 for his periodic behavior in the first quarter; Equipment Operator Sun participated in multiple technical drills, with a converted baseline data of 80.0; Assistant Engineer Li completed the basic training hours, with a baseline data of 40.0.

[0064] Regarding employee care scenarios: The system identified that Engineer Wang organized four large-scale volunteer activities. Based on the built-in algorithm (each activity is scored 45.0), the platform accurately calculated and obtained Engineer Wang's normalized baseline data for the first quarter as 180.0.

[0065] For the scenario of democratic suggestions: The system reads the joint proposal data of Engineer Zhou and Engineer Zhao, and directly inputs the objective status of their company-level adoption into the feature function preset in step 201. The feature function outputs the baseline performance value of the combination as 95.0 according to the status level.

[0066] To ensure data integrity during the next step of profile aggregation and avoid null values ​​that could cause algorithm errors, the system also supplemented basic seeding data across scenarios for all union members at this stage. The system retrieved data from the backend and found that although Mr. Zhang did not organize any activities in the first quarter, he participated in one basic union snow removal volunteer service, and thus received a default base score of 50.0 for him in the employee care scenario. Similarly, Mr. Wang participated in a basic safety skills quiz that was mandatory for all employees, and the system gave him a default base score of 85.0 for him in the skills enhancement scenario.

[0067] 203. The contribution index module is used to obtain the dynamic contribution index of each union member in each management scenario. Based on cooperative game theory and Monte Carlo simulation, it processes standardized data using the corresponding scenario-specific feature function to obtain the dynamic contribution index of the union member in the current scenario.

[0068] Specifically, to calculate the contribution index of the current union member in the current management scenario, the number of iterations of the Monte Carlo simulation is set to form simulation configuration parameters. Based on the simulation configuration parameters, a random permutation of the entire set of union members is obtained for a specified number of times, forming a set of random permutation sequences. For each permutation in the set of random permutation sequences, based on standardized data and the corresponding scenario-based feature function, the change in scenario performance value brought about by the current union member joining the subset of union members in front of it is calculated, obtaining the marginal contribution value of a single simulation. All simulated marginal contribution values ​​constitute a marginal contribution set. The arithmetic mean of all marginal contribution values ​​in the marginal contribution set is calculated to obtain the dynamic contribution index of the union member in the current management scenario.

[0069] Furthermore, based on standardized data, the basic performance ranking of all union members under the current management scenario is calculated to obtain an initial ranking sequence of union members. When obtaining each random arrangement, adjacent union members in the initial ranking sequence are swapped according to a preset perturbation probability to form a basic perturbation arrangement. A global shuffle operation is applied to the basic perturbation arrangement to ensure that all union member positions have the opportunity to be redistributed, resulting in a final single arrangement that meets the randomness requirement. The process of obtaining the basic perturbation arrangement and applying the global shuffle operation is repeated until a set of random arrangement sequences is obtained, where each final single arrangement is used as the union member order basis for the marginal contribution calculation.

[0070] It should be noted that, taking the skills enhancement and mentoring scenario in the first quarter as an example, the workshop technical improvement mutual assistance group composed of Engineer Zhang, Engineer Sun, and Engineer Li. The system retrieves the standardized basic performance values ​​of the three individuals obtained in step 202: Senior Engineer Zhang ( Equipment operator Sun ( Assistant Engineer Li () ).

[0071] To balance computational accuracy with cloud computing power consumption, the system set the number of Monte Carlo simulation iterations to N=10000. Subsequently, the system completely randomly arranged all union members in this scenario, generating an initial random sequence. To better reflect the actual technical guidance relationships within the enterprise (experts typically do not intervene as beginners), the cloud platform introduced an adaptive perturbation probability for swapping adjacent union member positions, correcting the initial random sequence to obtain an effective single permutation that satisfies both randomness requirements and aligns with business logic.

[0072] Completely random queuing violates the actual mentoring logic within enterprises (i.e., experts typically intervene at critical junctures), while completely fixed queuing fails to generate the marginal difference required by game theory. Therefore, the cloud platform employs an adaptive formula for calculating the probability of perturbations caused by swapping adjacent union member positions:

[0073] in, The probability (dimensionless, range) of two adjacent union members swapping positions in the current simulation. ); The system's preset basic disturbance constant (set to 0.5); , V represents the base performance value of two adjacent players; avg The average baseline performance of the union members in this group (V in this example) avg =(180+80+40) / 3=100).

[0074] The closer two players are in basic ability, the greater the likelihood of them switching roles; for two players with vastly different abilities (Zhang and Li), the probability of them swapping roles decays exponentially. To smoothly implement the business constraint that the greater the ability gap, the lower the swap probability, the system employs an exponential decay function. Compared to linear functions, exponential functions ensure that the probability calculation result is always positive, and that the probability approaches zero infinitely when the performance difference is extremely large, without exhibiting negative anomalies, making the entire mathematical model more robust.

[0075] In a certain simulation, the system evaluates the probability of Sun (80) and Li (40) exchanging positions: |80-40| / 100=0.4, substituting into the formula, we get... (i.e., 33.5%). Since the difference was small, the system's pseudo-random number triggered a swap, and the sequence became [Zhang Gong, Li Gong, Sun Gong]. Subsequently, the probability of Zhang Gong (180) versus Li Gong (40) was evaluated: |180-40| / 100=1.4, with a probability of only [33.5%]. (12.3%), no swap was triggered. After local perturbation combined with a global light shuffle, the system obtained the legal single permutation: [Sun Gong, Li Gong, Zhang Gong].

[0076] In the sequence [Sun Gong, Li Gong, Zhang Gong], the system tracks Zhang Gong's marginal contribution.

[0077] The performance of the first two subsets is calculated as follows: Sun joined first, with a performance score of 80; Li joined later. Since there are no strong association labels (mentor / apprentice or cross-disciplinary collaboration) between Sun and Li in the system database, the feature function first determines it as basic collaboration. After adding their respective base scores (80 and 40), it is multiplied by a preset base collaboration coefficient (1.05), resulting in a performance score of 126 for the first two subsets. If strong association labels such as mentor / apprentice exist, a higher-order collaboration amplification reward is triggered, and the score will be much higher.

[0078] Next, the target union member, Engineer Zhang, joined: Ranked third, Engineer Zhang not only had a base score of 180, but the feature function also keenly detected that his mentor-apprentice pairing with Engineer Li was still valid. The underlying logic instantly triggered a higher-order collaborative amplification reward, calculating that the total performance value of the complete three-person subset (Engineer Sun + Engineer Li + Engineer Zhang) soared to 450.

[0079] At this point, the system subtracts the performance value of the preceding subset (450-120=330) from the total performance value, accurately extracting Zhang Gong's marginal contribution value of 330 in this simulation.

[0080] The cloud platform automatically repeated the above simulation 10,000 times. Zhang Gong's marginal contribution value changed continuously depending on his appearance order and the teammates before him (the marginal value was 180 when he appeared first, and as high as 330 when he appeared last and triggered collaboration). The system calculated the arithmetic mean of these 10,000 marginal contribution values, set the expected value to be 242.6, and obtained the initial dynamic contribution index of the guild member in the current scenario. The initial dynamic contribution indices of all guild members in all scenarios were summarized, and the Min-Max standardization or Z-Score standardization method was used to uniformly map them to the percentage range of [0,100] to obtain the standard dynamic contribution index (Zhang Gong's 242.6 was normalized to 96.0), which was used for subsequent profile aggregation.

[0081] 204. The comprehensive profile module is used to aggregate the dynamic contribution index of each union member under all management scenarios to form a comprehensive contribution profile of the union member.

[0082] Specifically, the dynamic contribution index of each union member across all management scenarios is aggregated to form a multi-scenario contribution data set for union members. Based on preset business importance rules, an aggregation weight coefficient is configured for each management scenario to form scenario weight configuration parameters. For each union member, the dynamic contribution index of each scenario in the multi-scenario contribution data set is weighted and calculated according to the scenario weight configuration parameters to obtain the weighted comprehensive contribution value of the union member. The weighted comprehensive contribution value of each union member and its dynamic contribution index in each scenario are integrated to construct a comprehensive contribution profile of union members that includes quantitative contribution value and multi-dimensional performance details.

[0083] Furthermore, for each management scenario, the numerical distribution of the dynamic contribution index of all union members in that scenario is analyzed to identify its dispersion and central tendency, thus obtaining the contribution distribution characteristic data for that scenario. Based on the contribution distribution characteristic data, the ability of the contribution index to distinguish the differences among union members under different management scenarios is evaluated, and the discrimination coefficient of each scenario is calculated. According to the discrimination coefficient and the preset business priority rules, an initial aggregation weight is calculated for each management scenario. Within a predetermined observation period, the changes in the discrimination coefficient of the contribution index under each management scenario are continuously tracked, and the initial aggregation weight is dynamically corrected according to the changing trend to obtain and apply the scenario weight configuration parameters.

[0084] It should be noted that the core task of step 204 is dynamic weight allocation and multi-dimensional profile synthesis based on data distribution characteristics. Unlike traditional fixed-ratio KPI assessments, the cloud platform introduces discrete trend analysis from statistics, enabling the system to automatically identify which union work best reflects the real gaps between employees, thereby giving them a more reasonable voice.

[0085] At the end of the quarter, the cloud platform summarizes the dynamic contribution indices of all union members across three scenarios. Combining the derivations in steps 202 and 203, the system extracts the complete data sets for Senior Engineer Zhang and Enthusiastic Employee Wang: Zhang (Skill Improvement Index 242.6, Employee Care Index 50.0, Democratic Suggestion Index 120.0); Wang (Skill Improvement Index 85.0, Employee Care Index 210.0, Democratic Suggestion Index 90.0). Note that Zhang's Care Index of 50.0 and Wang's Skill Index of 85.0 both originate from the cross-scenario basic behavioral baseline data automatically obtained by the system in step 202.

[0086] While merging individual scores, the system conducted variance and standard deviation analysis on the overall scores of all 3,000 employees across the factory in three scenarios. The system found that the data in the skills enhancement scenario exhibited extremely high long-tail dispersion (extremely large Standard Deviation), with a few top-performing technical staff scoring very high, while most ordinary employees scored relatively low. Conversely, in the employee care scenario, due to widespread participation in donations and snow removal, scores were highly concentrated between 40 and 60 points, with extremely low dispersion (minimal standard deviation).

[0087] The cloud platform's algorithm engine determined that scenarios with higher dispersion have a stronger ability to identify and differentiate talent, and should therefore be assigned a higher discrimination coefficient. Based on the variance assessment results and combined with the union's pre-set business priorities for promoting production through technology this year, the system automatically obtained dynamic weight configuration parameters for the three management scenarios this quarter through weighted matrix calculation: the skill enhancement scenario with the highest discrimination coefficient was set to 50%; the democratic suggestion scenario with a medium discrimination coefficient was set to 30%; and the employee care scenario with the lowest discrimination coefficient was set to 20%. This dynamic weight mechanism effectively prevents the high scores of low-threshold scenarios that are available to everyone from diluting the high scores of high-threshold scenarios in core business.

[0088] After establishing the dynamic weights, the cloud platform performed a rigorous mathematical weighted synthesis of the various indices for Engineer Zhang and Engineer Wang: The calculation path for Zhang Gong's weighted comprehensive contribution value is: (96.0×0.5)+(50.0×0.2)+(95.0×0.3)=48.0+10.0+28.5=86.5 points (here it is simultaneously assumed that his democratic suggestion index is 120, which is normalized to 95.0).

[0089] The calculation path for Wang Gong's weighted comprehensive contribution value is: (85.0×0.5)+(210.0×0.2)+(90.0×0.3)=42.5+42.0+27.0=111.5 points.

[0090] These numerical values ​​are then converted into a comprehensive performance profile. The system uses threshold judgment to record Zhang's comprehensive score of 86.5, consistently ranking in the top 3% of the entire factory. Due to an extreme peak in his skill index, the system automatically assigns him semantic tags as a technical mentor and leader, and a core think tank member, resulting in a sharp radar chart biased towards technology. Wang's comprehensive score is 111.5, ranking in the top 20% of the entire factory. The system identifies his rare high score of over 200 in the employee care category, and automatically adds the distinctive tags of "Union Warm-hearted Ambassador" and "Highly Active Volunteer Pioneer," resulting in a service-oriented radar chart biased towards humanistic care.

[0091] 205. Management Operation Module, used to drive the cloud platform to automatically complete at least one of the following management operations based on the comprehensive contribution profile of union members: differentiated allocation of welfare resources, targeted scheduling of activity resources, and ranking and granting of awards.

[0092] Specifically, based on the type of management operation to be performed, corresponding contribution data is extracted from the comprehensive contribution profile of union members to obtain management decision input data; the management decision input data is input into a preset business decision rule set, and calculations and matching are performed according to the rules to obtain a set of management parameters including specific allocation quotas, scheduling priorities or sorting results; based on the set of management parameters, a set of executable instructions is obtained to drive the corresponding business subsystem in the cloud platform to perform specific operations; the set of executable instructions is executed to complete resource allocation, scheduling or qualification granting operations, and update the relevant member account status and system resource records.

[0093] It should be noted that at the quarterly summary point, the cloud platform triggers an automatic inspection mechanism. The system retrieves the status of the employee welfare resource pool (current inventory includes: 10 high-level intelligent health and wellness check-up slots, 50 special outdoor sports protective gear packages, and a number of advanced training slots and performance evaluation quotas).

[0094] The profile data of Engineer Zhang (overall score 86.5, Top 3% in the factory, tagged: Technical Leader) and Engineer Wang (overall score 111.5, Top 20% in the factory, extremely high in employee care, tagged: Union Warm-hearted Ambassador) were extracted and fed into the decision engine as input data for management decisions. The engine has built-in strict matching rules for three dimensions: Award criteria: Those who rank in the top 5% of the entire factory and possess a leader-level hardcore label will be directly locked in first place for the selection of high-level honors.

[0095] Welfare tier allocation rules: The top 5% in the overall ranking will trigger the highest level of health benefits; those who break through into the top 2% in a single index and have a care tag will trigger specific material rewards.

[0096] Resource allocation rules: Employees with outstanding technical skills will be given priority access to external technical training resources; employees with outstanding organizational service skills will have their account management privileges in the internal public welfare system automatically upgraded.

[0097] The decision engine completes data comparison and rule matching for thousands of users and obtains underlying executable instructions (APIPayloads), which are then distributed to various business subsystems for execution.

[0098] For Mr. Zhang, the system performed three concurrent operations in the background: First, through the API interface of the company's internal OA system, the system automatically called and filled in the "Quarterly Technical Pioneer Evaluation Application Form", listing Mr. Zhang as the first candidate, directly skipping the workshop-level review, and pushing it to the union chairman's terminal for approval; Second, it sent an instruction to the big health subsystem to automatically upgrade Mr. Zhang's employee ID's annual basic physical examination permission to a high-level intelligent health and wellness physical examination package, and pushed a unique electronic voucher through WeChat; Third, through the training scheduling module, it locked Mr. Zhang's VIP seat at the National Craftsmen Cross-Provincial Exchange Conference to be held next month, and triggered the travel system's booking process.

[0099] For Mr. Wang, although his total score did not trigger the top-tier comprehensive benefits, the system accurately captured his exceptionally high score in the care category and executed two targeted operations: First, it triggered the warehouse outbound API of the company's internal mall, automatically generating a shipping label to send a customized outdoor sports protective gear gift pack directly to Mr. Wang's workstation by the logistics department as a precise reward for his long-term organization of volunteer activities; Second, in the internal volunteer permission management module, through database SQL commands, Mr. Wang's account role was upgraded from ordinary participant to corporate-level public welfare project manager, and an initial quarterly activity approval quota of 5,000 yuan was automatically allocated to his account.

[0100] The comparison table between management execution automation and resource scheduling is shown in Table 2 below: Table 2

[0101] Through this highly digitalized operation, smart trade unions eliminate the injustices caused by human intervention, ensuring that every drop of sweat can be transformed into resource returns that best suit individual needs through precise data calculation.

[0102] 206. The Operation Feedback Module is used to collect feedback data from union members after completing management operations, and associate the feedback data with the corresponding management parameter set to obtain operation feedback records. Based on the operation feedback records, the module analyzes the correlation between union member contribution index and feedback satisfaction under different management scenarios to obtain scenario weight adjustment suggestion data. A configuration interface is provided in the cloud platform to receive confirmation or modification of the scenario weight adjustment suggestion data from the administrator, and update the scenario weight configuration parameters accordingly. The updated scenario weight configuration parameters are applied when performing contribution index calculation and management operations in subsequent cycles.

[0103] Furthermore, for each management operation, based on the satisfaction feedback from union members in the operation feedback record, the deviation between the actual management result and the union members' expectations is calculated to obtain satisfaction deviation data. The satisfaction deviation data is then correlated with the corresponding dynamic contribution index used when the operation was performed to calculate the utility coefficient of the contribution index, forming a correlation analysis result. Based on the correlation analysis result, the effectiveness of the dynamic contribution index of each management scenario under the current decision-making rules is evaluated, and the required adjustment value for the weight of each scenario is calculated. By combining the evaluation results of all management operations, suggested data for adjusting the scenario weights to guide the correction of the relative importance of each scenario is obtained.

[0104] It should be noted that within one month of the implementation of all health check packages, gift packs, honorary nominations, and managerial authority distribution in the first quarter (Q1), the cloud platform automatically pushed anonymous Likert Scale questionnaires to both covered and uncovered core union members via internal communication software. The system successfully collected the data and deeply mapped and correlated it with underlying management parameters.

[0105] Data shows that Mr. Zhang highly appreciates the advanced medical check-up and inter-provincial exchange services he received, giving him a satisfaction rating of 4.9 out of 5; Mr. Wang also achieved a satisfaction rating of 4.8 for the protective gear package and the administrator's privileges. Up to this point, the system's allocation seems perfect.

[0106] Meanwhile, feedback from the non-rewarded group was tracked: a group of highly active volunteers, whose individual performance in the employee care scenario ranked in the top 5% to 15% of the entire factory, experienced an unusually low average satisfaction score of only 3.2. The system retrieved historical expectation data from these employees and calculated through a mathematical model that their satisfaction deviation was extremely high. After tracing the algorithm's origins, the root cause was discovered: in the Q1 preset, the weight ceiling for the employee care scenario was set at a relatively low 20%. This meant that even though these volunteers achieved a high score of 180 in the care scenario, after the 20% weight reduction, their weighted comprehensive score still could not break through the factory's reward threshold, resulting in a collective sense of disappointment that despite their hard work, they were not recognized by the overall system.

[0107] The cloud platform's deep learning engine quickly intervened, performing Pearson correlation analysis on the collected deviation data and the corresponding contribution index to obtain the index utility coefficient.

[0108] In the skills enhancement scenario, the system found that high scorers received high rewards as desired, with no serious complaints. The calculated utility coefficient between the contribution index and the final incentive result for this scenario was as high as 0.92, indicating that the current 50% high weighting is extremely effective and credible in this business line.

[0109] In contrast, examining the employee care scenario, the system discovered a severe disconnect between the actual allocation results and the immense enthusiasm of employees. The calculated utility coefficient for this scenario plummeted to 0.65. The system concluded that while individual indices could accurately identify talent, the excessively low overall weight suppressed organizational cohesion among non-technical employees, indicating a significant imbalance in the weighting of this business line.

[0110] Based on the above utility coefficient assessment, the cloud platform algorithm engine automatically obtained scenario weight adjustment suggestions for the second quarter (Q2): that is, to moderately reduce the monopolistic proportion of the skill dimension and significantly enhance the basic status of employee care.

[0111] When the union's top administrator logged into the cloud platform's backend control center, the system proactively displayed a visual dashboard with global algorithm optimization suggestions. The dashboard used an intuitive data heatmap to list the pain points identified in Q1 and provided clear recommendations for new weighting ratios in Q2: it suggested increasing the weight of employee care from 20% to 25%, slightly adjusting skills enhancement from 50% to 45%, and keeping democratic suggestions unchanged at 30%. Considering the recent direction of the company's culture development (strongly advocating for employee mutual assistance and ESG principles), the administrator deemed the algorithm's suggestions highly forward-thinking and reasonable, and therefore clicked to confirm and override the changes.

[0112] As the second quarter drew to a close, when the system re-executed steps 204 (aggregated profiles) and 205 (resource allocation), the old weighting parameters were completely obsolete. The cloud platform will strictly apply the updated (45%, 30%, 25%) new configuration for weighting all users.

[0113] The dynamic evaluation and correction ecosystem comparison table for management scenario weights is shown in Table 3 below: Table 3

[0114] Through the above process, this smart trade union system built on the cloud platform is no longer a rigid program, but a digital life form with perception, calculation and repair capabilities. It uses data to protect the absolute fairness of the evaluation, while retaining the warmth and flexibility that trade union management should have with its keen feedback mechanism.

[0115] Figure 3 This is a schematic diagram of a cloud-based smart trade union management device 300 provided in an embodiment of the present invention. The cloud-based smart trade union management device 300 can vary significantly due to different configurations or performance. It may include one or more central processing units (CPUs) 310 (e.g., one or more processors) and a memory 320, and one or more storage media 330 (e.g., one or more mass storage devices) for storing applications 333 or data 332. The memory 320 and storage media 330 can be temporary or persistent storage. The program stored in the storage media 330 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the cloud-based smart trade union management device 300. Furthermore, the processor 310 may be configured to communicate with the storage media 330 and execute the series of instruction operations in the storage media 330 on the cloud-based smart trade union management device 300.

[0116] The cloud-based smart trade union management device 300 may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input / output interfaces 360, and / or one or more operating systems 331, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 3 The illustrated structure of the cloud-based smart union management device does not constitute a limitation on the cloud-based smart union management device. It may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0117] The present invention also provides a cloud-based smart trade union management device, which includes a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor performs the steps of the cloud-based smart trade union management method in the above embodiments.

[0118] The present invention also provides a cloud-based smart trade union management device, which includes a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor performs the steps of the cloud-based smart trade union management system described in the above embodiments.

[0119] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the cloud-based smart trade union management system.

[0120] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0121] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, 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 steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0122] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A cloud-based intelligent trade union management system, characterized in that, include: The scenario definition module is used to define multiple independent union management scenarios in the cloud platform and define a scenario-based feature function for each management scenario; The data processing module is used to continuously collect behavioral data from all union members and process the behavioral data into standardized data. The contribution index module is used to process the standardized data for each union member in each management scenario using the corresponding scenario-based feature function to obtain the dynamic contribution index of the union member in the current scenario. The comprehensive profile module is used to aggregate the dynamic contribution index of each union member under all management scenarios to form a comprehensive contribution profile of the union member. The management operation module is used to drive the cloud platform to complete at least one management operation based on the comprehensive contribution profile of the union members, including the differentiated allocation of welfare resources, the targeted scheduling of activity resources, and the ranking and granting of awards.

2. The cloud-based smart trade union management system according to claim 1, characterized in that, include: Define multiple independent business dimensions of trade union management to form multiple independent management scenarios corresponding to different management objectives; Define a scenario-based feature function for each of the management scenarios. All scenario-based feature functions constitute a function set, wherein each function is configured to receive data from a subset of union members and output a quantitative value that represents the overall performance of the subset in the scenario. Based on the evaluation objectives of each management scenario, the specific behavioral data type required for the calculation of each scenario-based feature function is configured, and a mapping relationship between the behavioral data type and the scenario-based feature function is established.

3. The cloud-based smart trade union management system according to claim 2, characterized in that, include: Collect raw behavioral records of all union members in various activities and interactions to form a raw behavioral data set; Based on the established mapping relationship, the behavioral data necessary for the calculation of the corresponding scenario-based feature function is selected from the original behavioral data set for each management scenario, and the dataset to be processed for each scenario is obtained. The behavioral data in each of the datasets to be processed are processed to obtain a standardized data set for each scenario; The numerical data in each of the aforementioned standardized datasets are processed to obtain standardized data that can be directly used by the scenario-specific feature functions.

4. The cloud-based smart trade union management system according to claim 3, characterized in that, From the business database of the cloud platform, we extract participation records of union members participating in various union activities, the content of formal proposals and suggestions submitted by union members, and the initiation and completion records of mutual assistance behaviors among union members to obtain categorized behavior records. Each record in the typified behavior record is marked with its occurrence timestamp, and quantifiable evaluation indicators are identified to form quantified behavior data with time-series labels; Based on time-stamped quantitative behavioral data, the periodic statistical values ​​of each union member in each type of behavior are calculated using a preset statistical period as the unit, thus obtaining the baseline data of the periodic behavior of union members. Based on the target time window corresponding to the current calculation of the contribution index, statistical values ​​of all relevant periods covering the time window are aggregated from the baseline data of the periodic behavior of union members to form the original behavior data set.

5. The cloud-based smart trade union management system according to claim 3, characterized in that, include: To calculate the contribution index of current union members under the current management scenario, the number of iterations of the Monte Carlo simulation is set to form simulation configuration parameters; Based on the simulation configuration parameters, a random permutation of the complete set of union members is obtained a specified number of times, forming a set of random permutation sequences; For each permutation in the set of random permutations, based on the standardized data and the corresponding scenario-based feature function, the change in scenario performance value brought about by the current union member joining the subset of union members in front of him is calculated to obtain the marginal contribution value of a single simulation. All simulated marginal contribution values ​​constitute the marginal contribution set. The dynamic contribution index of the union member under the current management scenario is obtained by calculating all marginal contribution values ​​in the marginal contribution set.

6. The cloud-based smart trade union management system according to claim 5, characterized in that, Based on the standardized data, the basic performance ranking of all union members under the current management scenario is calculated to obtain the initial ranking sequence of union members. When each random arrangement is obtained, the positions of adjacent union members in the initial sorting sequence of union members are swapped according to the preset perturbation probability to form a basic perturbation arrangement; A global shuffle operation is applied to the basic perturbation arrangement to ensure that all union member positions have the opportunity to be redistributed, resulting in a final single arrangement that meets the randomness requirement. Repeat the basic perturbation arrangement and apply the global shuffle operation until the set of random arrangement sequences is obtained, wherein each final single arrangement obtained is used as the basis for the union member order on which the marginal contribution calculation depends.

7. The cloud-based smart trade union management system according to claim 6, characterized in that, The preset perturbation probability is obtained through the following adaptive calculation formula: ； in, This represents the probability that two adjacent union members will swap positions in the current simulation. The basic disturbance constant preset for the system; , V represents the base performance value of two adjacent players; avg This represents the average basic performance of the union members in this group.

8. The cloud-based smart trade union management system according to claim 5, characterized in that, The cloud platform automatically performs management operations, including: Based on the type of management operation to be performed, the corresponding contribution data is extracted from the comprehensive contribution profile of the union members to obtain management decision input data; The management decision input data is input into a preset business decision rule set to obtain a set of management parameters, including specific allocation quotas, scheduling priorities, or sorting results; Based on the set of management parameters, a set of executable instructions is obtained to drive the corresponding business subsystem in the cloud platform to perform specific operations.

9. The cloud-based smart trade union management system according to claim 1, characterized in that, It also includes an operation feedback module: Collect feedback data from union members regarding the management operations, and associate the feedback data with the management parameter set corresponding to the operation to obtain the operation feedback record; Based on the operation feedback records, the correlation between the union member contribution index and feedback satisfaction under different management scenarios is analyzed to obtain scenario weight adjustment suggestions. A configuration interface is provided in the cloud platform to receive confirmation or modification of the scenario weight adjustment suggestions from the administrator, and update the scenario weight configuration parameters accordingly.

10. The cloud-based smart trade union management system according to claim 9, characterized in that, The analysis of the correlation between union member contribution index and feedback satisfaction under different management scenarios includes: For each management operation, based on the satisfaction feedback of union members in the operation feedback record, the deviation between the actual management result and the expectations of union members is calculated to obtain the satisfaction deviation data. The satisfaction deviation data is correlated with the corresponding dynamic contribution index used when performing the operation, and the utility coefficient of the contribution index is calculated to form the correlation analysis results. Based on the correlation analysis results, the effectiveness of the dynamic contribution index of each management scenario under the current decision-making rules is evaluated, and the adjustment change value required for the weight of each scenario is calculated. The evaluation results of all management operations are used to obtain suggested data for adjusting the relative importance of each scenario.

Images