Human resource allocation system for construction enterprise project groups based on multi-objective optimization

Abstract

The present invention relates to the technical field of human resource management methods and discloses a human resource allocation system for construction enterprise project groups based on multi-objective optimization. The system comprises: a model construction module for constructing a multi-objective optimization model that maximizes the overall person-job matching of the project group, optimizes the age structure of team members, and minimizes human resource costs; an information acquisition module for acquiring employee information and job information; a matching quantification module for quantifying person-job matching based on job competency models and personnel portraits, and evaluating the rationality of the team member age structure using an employee marginal contribution rate curve; and a solution generation module for solving the multi-objective optimization model using an improved NSGA-II algorithm to obtain and output a human resource allocation solution. The present invention aims to provide a multi-objective optimization system that can simultaneously consider person-job matching, team age structure, and human resource costs, so as to improve the scientific nature and effectiveness of human resource allocation in construction enterprises.

Description

Technical Field

[0001] The present invention relates to the technical field of human resource management methods, and in particular to a human resource configuration system for a construction enterprise project group based on multi-objective optimization. Background Art

[0002] Human resources are one of the most critical production factors for construction companies, and their rational allocation is directly related to the company's operational efficiency and project success rate. This is especially true in the project management of large-scale construction companies, where multiple projects are being carried out simultaneously, making human resource allocation more complex and placing higher demands on the company's management capabilities.

[0003] At present, construction companies generally have the following problems in their human resource allocation systems:

[0004] Inefficient human resource allocation: Traditional human resource allocation systems rely heavily on empirical judgment and manual analysis, lacking scientific quantitative metrics processing and system optimization algorithms. Project managers often make personnel assignments based on subjective experience or simple qualification matching, resulting in poor matchmaking and significant resource waste.

[0005] Lack of comprehensive consideration of multiple objectives: Existing human resource allocation systems often only contain single-objective processing modules (such as cost minimization modules or resource utilization maximization modules), and lack comprehensive processing modules that can balance multiple objectives such as job matching, team structure optimization and cost control, resulting in poor overall allocation results.

[0006] Intense competition for project resources: In a project management environment, subprojects are run in parallel, preventing employees from being reused across subprojects, exacerbating competition for human resources. The existing system lacks an effective resource prioritization module and is unable to meet the differentiated needs of multiple projects.

[0007] Unreasonable team structure: The existing system ignores the impact of the team member age structure assessment module on project performance, resulting in the team structure being too old or too young, which is not conducive to the overall performance of the team.

[0008] Low algorithm processing efficiency: For complex multi-objective optimization problems, the optimization algorithm processing modules in existing systems often have high computational complexity and low solution efficiency, making it difficult to meet actual application needs. Summary of the Invention

[0009] In order to solve the problems existing in the above-mentioned prior art, the present invention aims to provide a multi-objective optimization system that can simultaneously consider the person-job matching, team age structure and human resource costs, and achieve rapid solution through an efficient algorithm to improve the scientificity and effectiveness of human resource allocation in construction enterprises.

[0010] In order to achieve the above-mentioned object of the invention, the technical solution provided by the present invention includes:

[0011] The human resource allocation system for construction enterprise project groups based on multi-objective optimization includes:

[0012] The model building module is used to build a multi-objective optimization model that includes maximizing the overall person-job matching of the project group, optimizing the age structure of the team members, and minimizing human resource costs, and setting allocation constraints for employees and positions;

[0013] An information acquisition module, connected to the model building module by signals, for acquiring employee information and position information, and transmitting the acquired information to the model building module and the matching quantification module;

[0014] A matching quantification module is connected to the information acquisition module and the model construction module by signal, and is used to receive the employee information and position information transmitted by the information acquisition module, quantify the person-position matching degree based on the position competency model and personnel portrait, and use the employee marginal contribution rate curve to evaluate the rationality of the team's personnel age structure, and transmit the quantification result signal to the model construction module and the solution generation module;

[0015] The solution generation module is connected to the model construction module and the matching quantization module signal, and is used to receive the multi-objective optimization model transmitted by the model construction module and the quantitative result signal transmitted by the matching quantization module, use the improved NSGA-Ⅱ algorithm to solve the multi-objective optimization model, obtain the human resource allocation solution and output it.

[0016] Preferably, the employee information includes the employee's skill level, work experience, educational background, performance, professional qualification certificate and age information; the position information includes the job responsibilities, skill requirements, experience requirements, educational requirements, qualification certificate requirements and age structure requirements.

[0017] Preferably, the matching quantification module includes:

[0018] A competency model building unit, configured to build a competency model for each position based on the position information and determine the weight of each competency indicator;

[0019] A personnel profile building unit, configured to build a personnel profile of the employee based on the employee information and quantify the scores of various competency indicators;

[0020] A matching calculation unit is used to calculate the person-job matching score of each employee and each job based on the competency model of the job and the personnel portrait of the employee.

[0021] Preferably, the matching quantification module further includes an age structure assessment unit, which includes:

[0022] The contribution rate curve fitting subunit is used to fit the employee marginal contribution rate curve based on the employee's historical performance data and age information;

[0023] a contribution level determination subunit, configured to determine the expected contribution levels of employees of different age groups based on the employee marginal contribution rate curve;

[0024] The team contribution calculation subunit is used to calculate the expected contribution of the team as a whole based on the age distribution of the current team members and the expected contribution level of employees in each age group;

[0025] The structural rationality assessment sub-unit is used to compare the expected contribution of the team as a whole with the expected contribution under the preset reasonable age structure, and to evaluate the rationality of the current team age structure.

[0026] Preferably, the method in which the solution generation module uses the improved NSGA-II algorithm to solve the multi-objective optimization model, obtains and outputs the human resource allocation solution includes:

[0027] S101. Perform double-layer segmented encoding on chromosomes, where the chromosomes are used to simultaneously represent both personnel-project allocation information and personnel-position allocation information. The first layer of chromosomes represents the project to which the personnel are assigned, and the second layer of chromosomes represents the degree of match between the personnel and the position, expected salary, and age.

[0028] S102. Use non-dominated sorting based on early rejection strategy to accelerate the sorting process of individuals in the population;

[0029] S103. Performing individual selection based on an improved niche strategy, wherein the niche radius is dynamically determined based on the Euclidean distance between individuals in the population;

[0030] S104. Perform a crossover operation using a partial matching crossover operator to ensure the genetic uniqueness of the chromosome after the crossover; perform a single-point mutation operation on the chromosome and perform a deduplication process to ensure the genetic uniqueness of the chromosome after the mutation;

[0031] S105. Terminate the algorithm based on the number of iterations and the fitness value, and output the Pareto optimal solution set as the human resource allocation plan.

[0032] Preferably, the employee and position allocation constraints set by the model building module include:

[0033] Each employee is assigned to at most one position, while some employees are allowed to remain unassigned;

[0034] Each position must be assigned an employee;

[0035] A priority weight coefficient is set for each project, where the priority weight coefficient is a value between 0 and 1, and the sum of the priority weight coefficients of all projects is 1.

[0036] Preferably, the objective function of the multi-objective optimization model constructed by the model construction module includes:

[0037] S201. Maximize the overall project team's job matching , expressed as:

[0038] ;

[0039] in, Represents maximizing the overall person-job matching of the project group function, It represents the priority weight coefficient of project h, N is the total number of projects, n is the personnel number, and m is the position number; It represents the matching degree between position j of project h and person i; It is a 0-1 variable. When the value is 1, it means that person i has been assigned to position j.

[0040] S202. Optimize the age structure of team members , expressed as:

[0041] ;

[0042] in, Represents the maximum age structure of the team members function, It represents the evaluation value of the rationality of the age structure of the team members of project h;

[0043] S203. Minimize human resource costs , expressed as:

[0044] ;

[0045] in, Minimize human resource costs function, It represents the expected salary of person i for position j.

[0046] Preferably, the method for individual selection based on the improved niche strategy comprises:

[0047] S301. Calculate the Euclidean distance between individuals in a population:

[0048] ;

[0049] Where x and y represent two random individuals in the population. represents the Euclidean distance between x and y, xᵢ and yᵢ are the values ​​of the i-th objective function corresponding to individual x and individual y respectively;

[0050] S302. After calculating the Euclidean distance between each individual and other individuals, the average of the minimum Euclidean distance values ​​of all individuals is taken as the initial value of the microhabitat radius;

[0051] S303. Dynamically iterate and dynamically adjust the microhabitat radius. The microhabitat radius is gradually reduced according to the following formula:

[0052] ; represents the current microhabitat radius, is a constant; gs is the current number of iterations, gs max is the maximum number of iterations;

[0053] S304. When the Euclidean distance between two individuals x and y is less than the current microhabitat radius When , the fitness values ​​of the two individuals are compared, and the individual with lower fitness value is punished to reduce its probability of being selected.

[0054] Preferably, the process of calculating the job matching degree includes the following sub-steps:

[0055] S401. Analyze job attributes and determine the required employee capabilities. The set of positions requiring personnel in the project is denoted as J, and the set of available personnel for each position is denoted as I.

[0056] S402. Obtain the matching results of person i and position j on the tth qualification factor respectively , the matching result between person i and position j on the sth ability factor , and the competency evaluation value of person i in position j ;

[0057] S403. Based on the above evaluation results, calculate the job matching degree of person i in job j :

[0058] ;

[0059] in, is a piecewise function and ; Q is the penalty coefficient, is the weight of the ability factor.

[0060] Beneficial effects

[0061] 1. Improving the Accuracy of Personnel-Job Matching: This system accurately quantifies and matches employee capabilities with job requirements by building competency models and personnel profiles within the matching quantification module. This overcomes the subjectivity and arbitrariness inherent in traditional systems that rely on empirical judgment, making personnel allocation more scientific and rational. Specifically, the system's matching calculation unit simultaneously considers both the assessment of basic job requirements and the comprehensive assessment of competencies and qualities. It also prioritizes basic requirements through piecewise functions and penalty mechanisms, significantly improving the accuracy of person-job matching.

[0062] 2. Achieve multi-objective collaborative optimization: The system's model building module constructs a multi-objective optimization model based on job-person fit, team age structure, and salary expenditure. This model breaks through the limitations of traditional single-objective optimization systems and achieves multi-objective collaborative optimization. By setting differentiated priority weights for projects, it further improves the targetedness and flexibility of resource allocation, enabling enterprises to adjust resource allocation strategies based on actual needs.

[0063] 3. Improving Algorithm Solution Efficiency: The solution generation module of this system utilizes the NSGA-II algorithm based on a multi-strategy improvement. This significantly improves the algorithm's solution speed and result quality through the early rejection strategy of the sorting acceleration unit and the improved niche strategy of the individual selection unit. The dual-layer segmented encoding of the chromosome encoding unit enables the system to simultaneously represent both person-project and person-position allocation relationships. The partial matching crossover and single-point mutation operations of the crossover mutation unit ensure effective understanding, enabling optimization results to better guide actual human resource allocation decisions.

[0064] 4. Reduced Human Resource Costs: By minimizing the human resource cost unit within the model building module, this system effectively controls a company's human resource costs. Actual application cases demonstrate that, compared to traditional manual allocation systems, this system significantly reduces a company's human resource expenditures and improves its economic benefits, while ensuring a good match between people and jobs and a reasonable team structure.

[0065] 5. Improved Project Management Efficiency: This system improves the efficiency and scientific nature of human resource allocation within construction enterprise project groups, reduces the time and errors associated with manual decision-making, and makes project management more efficient. In practical applications, this system can quickly generate multiple alternative configuration options for decision-makers to choose from, greatly improving the flexibility and scientific nature of decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 This is a schematic diagram of the structure of a human resources allocation system for a construction enterprise project group based on multi-objective optimization provided in a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0067] like Figure 1As shown, this embodiment provides a human resource allocation system for construction enterprise project groups based on multi-objective optimization, including a model building module, an information acquisition module, a matching quantification module and a solution generation module. Signal connections are used between the modules to achieve information transmission and functional collaboration.

[0068] The model building module is used to construct a multi-objective optimization model that includes maximizing the overall matching degree of people and positions in the project group, optimizing the age structure of team members, and minimizing human resource costs, and setting allocation constraints for employees and positions. In the project group management of construction enterprises, human resource allocation needs to take into account multiple dimensions of goals at the same time. This module first determines three core objective function units: maximizing the overall matching degree of people and positions in the project group, optimizing the age structure of team members, and minimizing human resource costs. These three objective function units respectively perform multi-dimensional evaluations on human resource allocation from the perspectives of configuration efficiency, team structure, and economic cost. In some preferred embodiments, the objective functions of the multi-objective optimization model constructed by the model building module specifically include:

[0069] S201. Maximize the overall project team's job matching , expressed as:

[0070] ;

[0071] in, Represents maximizing the overall person-job matching of the project group function, It represents the priority weight coefficient of project h, N is the total number of projects, n is the personnel number, and m is the position number; It represents the matching degree between position j of project h and person i; It is a 0-1 variable. When the value is 1, it means that person i has been assigned to position j.

[0072] S202. Optimize the age structure of team members , expressed as:

[0073] ;

[0074] in, Represents the maximum age structure of the team members function, represents the evaluation value of the rationality of the age structure of the team members of project h. Specifically, the evaluation value of the rationality of the age structure of the team members is calculated by the age structure evaluation unit of the matching quantification module, and the age structure evaluation unit includes:

[0075] The contribution rate curve fitting subunit is used to fit the employee marginal contribution rate curve based on the employee's historical performance data and age information;

[0076] a contribution level determination subunit, configured to determine the expected contribution levels of employees of different age groups based on the employee marginal contribution rate curve;

[0077] The team contribution calculation subunit is used to calculate the expected contribution of the team as a whole based on the age distribution of the current team members and the expected contribution level of employees in each age group;

[0078] The structural rationality assessment sub-unit is used to compare the expected contribution of the team as a whole with the expected contribution under the preset reasonable age structure, and to evaluate the rationality of the current team age structure.

[0079] S203. Minimize human resource costs , expressed as:

[0080] ;

[0081] in, Minimize human resource costs function, It represents the expected salary of person i for position j.

[0082] To ensure rational resource allocation, the model construction module also sets employee and position allocation constraints. Specifically, each employee can be assigned to at most one position, while some employees can remain unassigned to reflect the flexible deployment needs of actual work; each position must be assigned an employee to ensure basic project operations; and each project is assigned a priority weight coefficient, ranging from 0 to 1, with the sum of all project priority weight coefficients equal to 1, ensuring that high-priority projects receive more high-quality resources. This multi-objective model construction effectively balances the three key objectives of employee-position matching, team structure, and cost control, providing a clear mathematical model signal foundation for the solution generation module.

[0083] The information acquisition module, signal-connected to the model building module, is used to obtain employee and position information and transmit this information to the model building module and the matching quantification module. This module, the data foundation for system implementation, comprehensively collects and organizes relevant information about employees and positions through multiple information channels, including interfacing with the enterprise human resources management system, accessing employee resume databases, integrating with performance evaluation record systems, and collecting data from departmental needs surveys.

[0084] The information acquisition module collects employee information, primarily including basic data such as professional skills, work experience, educational background, performance, professional qualifications, and age. Position information encompasses key elements such as job responsibilities, required skills, experience requirements, educational requirements, and qualification requirements. This detailed employee and position information forms the foundation for assessing the fit between individuals and positions and the rationality of the team's age structure. It also provides the necessary signal input for the subsequent calculations and processing in the match quantification module and the optimization decisions in the solution generation module.

[0085] In some preferred embodiments, the employee information collected by the information acquisition module can be further broken down into more detailed dimensions such as years of experience, project type, project size, and role held. Position information can further include more detailed descriptions of job responsibilities, the importance of various skills, and ideal age structure requirements. The information acquisition module transmits this complete information to other modules of the system via a signal interface, ensuring data integrity and accuracy for system operation.

[0086] The matching quantification module is connected to the information acquisition module and the model construction module by signal, and is used to receive the employee information and position information transmitted by the information acquisition module, quantify the person-position matching degree based on the position competency model and personnel portrait, and use the employee marginal contribution rate curve to evaluate the rationality of the team personnel age structure, and transmit the quantification result signal to the model construction module and the solution generation module.

[0087] In some preferred embodiments, the module includes a competency model construction unit, a personnel portrait construction unit, and a matching degree calculation unit. The competency model construction unit receives the position information signal transmitted by the information acquisition module, constructs a competency model for each position, and determines the weight of each competency indicator; the personnel portrait construction unit receives the employee information signal transmitted by the information acquisition module, constructs the employee's personnel portrait, and quantifies the score of each competency indicator; and the matching degree calculation unit calculates the person-job matching score for each employee and each position based on the signals output by the competency model construction unit and the personnel portrait construction unit.

[0088] The matching quantification module also includes an age structure assessment unit, which, through a series of subunits, works together to assess the rationality of the team's age structure. Specifically, the contribution rate curve fitting subunit fits the employee marginal contribution rate curve based on the employee's historical performance data and age information; the contribution level determination subunit determines the expected contribution level of employees of different age groups based on the fitted employee marginal contribution rate curve; the team contribution calculation subunit calculates the expected contribution of the team as a whole based on the current age distribution of team members and the expected contribution level of employees of each age group; and the structural rationality assessment subunit compares the expected contribution of the team as a whole with the expected contribution under a preset reasonable age structure to evaluate the rationality of the current team's age structure.

[0089] These functional units convert qualitative personnel and position information into quantitative indicator signals, providing precise input parameters for the model construction module and the solution generation module. In some preferred embodiments, the quantification of the person-position matching in the matching calculation unit can be divided into two processing links: basic position condition assessment and comprehensive ability and quality assessment. The basic condition assessment processing unit mainly examines whether the rigid requirements of the position, such as academic qualifications and certificates, are met, while the comprehensive ability and quality assessment processing unit considers the degree of match between employees in various ability dimensions and position requirements. The age structure assessment unit can identify 31-40 years old as the best contribution age group by constructing an inverted U-shaped employee marginal contribution rate curve, and evaluate the team's overall expected contribution value signal accordingly.

[0090] The solution generation module is connected to the model construction module and the matching quantization module signal, and is used to receive the multi-objective optimization model transmitted by the model construction module and the quantitative result signal transmitted by the matching quantization module, use the improved NSGA-Ⅱ algorithm to solve the multi-objective optimization model, obtain the human resource allocation solution and output it.

[0091] This module is a key link in the implementation of the system. By improving the non-dominated sorting genetic algorithm (NSGA-Ⅱ), it can efficiently solve the multi-objective optimization problem of human resource allocation in construction enterprise project groups. The core idea of ​​the improved NSGA-Ⅱ algorithm is to maintain population diversity through non-dominated sorting and congestion calculation, while evolving towards the Pareto optimal frontier. The implementation process of the algorithm includes the main steps of population initialization, fitness evaluation, non-dominated sorting, crossover and mutation, and elite strategy selection, and finally outputs a series of non-dominated solutions that balance the three objectives as optional human resource allocation plans. In some preferred embodiments, the solution generation module uses the improved NSGA-Ⅱ algorithm to solve the multi-objective optimization model, and the method of obtaining and outputting the human resource allocation plan includes:

[0092] S101. Perform double-layer segmented encoding on chromosomes, where the chromosomes are used to simultaneously represent both personnel-project allocation information and personnel-position allocation information. The first layer of chromosomes represents the project to which the personnel are assigned, and the second layer of chromosomes represents the degree of match between the personnel and the position, expected salary, and age.

[0093] S102. Use non-dominated sorting based on early rejection strategy to accelerate the sorting process of individuals in the population;

[0094] S103. Perform individual selection based on the improved niche strategy, wherein the niche radius is dynamically determined based on the Euclidean distance between individuals in the population. In some preferred embodiments, a method for performing individual selection based on the improved niche strategy is provided, comprising:

[0095] S301. Calculate the Euclidean distance between individuals in a population:

[0096] ;

[0097] Where x and y represent two random individuals in the population. represents the Euclidean distance between x and y, with xᵢ and yᵢ representing the value of the i-th objective function for individuals x and y, respectively. Unlike traditional genetic algorithms, which focus solely on individual fitness, this method calculates the Euclidean distance to comprehensively consider the distribution of individuals across multiple objective dimensions, providing an objective basis for subsequent niche formation. In practical implementations, spatial indexing data structures such as KD trees can be used to accelerate distance calculations to improve computational efficiency. This optimization can reduce the time complexity of distance calculations for large populations.

[0098] S302. After calculating the Euclidean distance between each individual and every other individual, the average of the minimum Euclidean distance values ​​of all individuals is taken as the initial value of the microhabitat radius. Compared with the fixed radius commonly used in the prior art, the adaptive initial radius strategy can improve the convergence speed and increase the number of optimal solutions found.

[0099] S303. Dynamically iterate and dynamically adjust the microhabitat radius. The microhabitat radius is gradually reduced according to the following formula:

[0100] ; represents the current microhabitat radius, is a constant; gs is the current number of iterations, gs max The core idea of ​​this step is to gradually reduce the niche radius as the optimization process progresses, achieving a natural transition from global exploration to local, refined search. In the early stages of the iteration, a larger niche radius helps maintain population diversity and promotes extensive exploration of the solution space. In the later stages of the iteration, the niche radius decreases, allowing the algorithm to focus more on refined search in promising areas, improving the quality of the solution. Controls the rate of radius contraction. The larger the β value, the faster the radius contracts. Conversely, the slower the contraction. In some preferred embodiments, Use a value between 0.2 and 0.8.

[0101] S304. When the Euclidean distance between two individuals x and y is less than the current microhabitat radius When the fitness values ​​of the two individuals are compared, the one with the lower fitness value is penalized, reducing its probability of selection. This distance-based penalty mechanism is based on the "competitive exclusion principle" in biological evolution, which states that competition for resources between similar individuals leads to the elimination of individuals with weaker fitness. In multi-objective optimization problems, this mechanism helps maintain population diversity and prevents individuals from excessively concentrating in the same area, effectively preventing the algorithm from falling into a local optimum.

[0102] S104. Perform a crossover operation using a partial matching crossover operator to ensure the genetic uniqueness of the chromosome after the crossover; perform a single-point mutation operation on the chromosome and perform a deduplication process to ensure the genetic uniqueness of the chromosome after the mutation;

[0103] S105. Terminate the algorithm based on the number of iterations and the fitness value, and output the Pareto optimal solution set as the human resource allocation plan.

[0104] It should be understood that the employee and position information provided by the Information Acquisition Module provides the data foundation for the precise calculations of the Matching Quantification Module. This enables companies to make personnel allocation decisions based on objective data rather than subjective experience, significantly improving the scientific nature and accuracy of human resource allocation. Employee information is a comprehensive description of a company's current human resources situation, encompassing six key dimensions. Skill level refers to an assessment of an employee's competency in a specific professional field, typically obtained through an internal skills assessment system or professional assessment. This includes professional technical skills, management skills, communication and coordination abilities, and is expressed using a quantitative scoring system (e.g., a 1-5 scale). Work experience encompasses not only an employee's overall years of work experience but also project experience in specific areas of the construction industry, such as project type, scale, and complexity, as well as specific roles and responsibilities assumed. This information is typically obtained from an employee's resume and historical project records. Educational background refers to an employee's highest degree, major, and continuing education, reflecting their knowledge base and learning ability. Performance is the evaluation of an employee's work performance based on the company's historical performance appraisal system, encompassing aspects such as work quality, work efficiency, innovation, and teamwork. Professional qualification certificates refer to various professional qualifications held by an employee related to the construction industry, such as Registered Constructor or Registered Structural Engineer, including certificate type, level, and validity period. Age information directly reflects an employee's actual age and serves as the foundation for assessing the rationality of a team's age structure. This employee information is collected and organized through channels such as the company's human resources management system, employee files, performance evaluation systems, and professional skills assessment platforms to form a comprehensive employee profile.

[0105] Job information is a systematic description of the requirements for each job within a construction company's projects, also encompassing six key dimensions. Responsibilities define the main work content, scope of responsibilities, and work objectives of a job, and are the core of the job definition. Skill requirements detail the various professional skills required for the position and their minimum level, such as design, construction management, and budgeting capabilities, typically using quantitative standards corresponding to employee skill level assessments. Experience requirements specify the minimum number of years required for the position and the experience required for specific project types, such as "requires more than five years of experience in large-scale public construction project management." Education requirements define the minimum educational background required for the position, including academic level and major requirements, such as "bachelor's degree or above in civil engineering or a related major." Qualification certificate requirements list the required professional qualifications for the position, including certificate type and level requirements. Age structure requirements consider the preferred age range for employees based on job characteristics. For example, senior management positions requiring extensive experience may prefer employees aged 35-45, while construction positions with higher physical demands may be more suitable for employees aged 25-35. These job information are usually determined jointly by project managers, human resources departments and business departments, and are formulated based on project characteristics, industry standards and actual corporate needs to form a systematic job description.

[0106] In this invention, the quantitative calculation of job-person matching and the assessment of the rationality of team age structure in the matching quantification module are key links in achieving efficient human resource allocation. It should be understood that by using scientific and quantitative methods to evaluate the rationality of job-person matching and team age structure, traditional personnel allocation that relies on subjective judgment can be transformed into optimized decisions based on objective data, thereby significantly improving allocation efficiency and quality. The matching quantification module of the system includes the following functional units and signal processing flow:

[0107] The competency model construction unit constructs competency models for each position based on received position information signals and determines the weights of each competency indicator. The signal processing process includes the following: First, based on the received position information signals, the core competency indicators for each position are identified through data processing algorithms using expert interviews and critical incident analysis. These indicators typically include professional knowledge, professional skills, work experience, management skills, communication and coordination skills, problem-solving skills, and innovation capabilities. Second, the mathematical processing unit uses the analytic hierarchy process to determine the weights of each competency indicator. This involves constructing a judgment matrix, determining the relative importance of each indicator based on expert scoring signals, calculating eigenvectors, and performing consistency checks to ultimately obtain weight value signals for each indicator. Finally, the position competency standards are established, with scoring criteria set for each indicator (usually using a 1-5 point scale). For example, for the "project management experience" indicator, the scales might be: 5 points: more than 10 years of large-scale project management experience; 4 points: 7-10 years of project management experience; 3 points: 4-6 years of project management experience; 2 points: 1-3 years of project management experience; and 1 point: less than 1 year of project management experience. Through this signal processing method, a complete job competency model signal including indicator system, weight system and scoring standard was established.

[0108] The Personnel Profile Construction Unit constructs a profile of each employee based on the received employee information signals and quantifies the scores for each competency indicator. First, it integrates data signals transmitted by the Information Acquisition Module from the enterprise's human resources management system, performance evaluation system, skills assessment platform, and employee profiles to establish a multi-dimensional employee information database. Second, for each employee, it extracts information items corresponding to job competency indicators, such as professional qualifications, educational background, work experience, and performance evaluations. Then, based on pre-set quantification rules, it converts the employee's performance on each competency indicator into a standardized score signal. For example, for the "Professional Knowledge" indicator, the score can be based on the employee's academic qualifications (PhD: 5 points, Master's: 4 points, Bachelor's: 3 points, etc.) and professional qualification certificate level; for the "Work Experience" indicator, the score can be quantified based on years of work experience and project experience. This process can be completed automatically by the data mining algorithm unit or combined with manual evaluation signal input, ultimately forming a complete employee competency profile signal.

[0109] The match calculation unit calculates the person-job match score for each employee and each position based on the signals output by the competency model construction unit and the personnel profile construction unit. Using a weighted summation model processing unit, the employee's score on each competency indicator is multiplied by the weight of that indicator for the specific position and the resulting sum is calculated. The calculated person-job match score typically ranges from 0 to 5, with a higher score indicating a better match. Finally, the match scores for all employees and all positions are organized into a person-job match matrix signal and transmitted to the model construction module and solution generation module as an important input signal for multi-objective optimization.

[0110] In some other preferred embodiments, a process for calculating the job matching degree is provided, which specifically includes the following sub-steps:

[0111] S401. Analyze job attributes and determine the required employee capabilities. The set of positions requiring personnel in the project is denoted as J, and the set of available personnel for each position is denoted as I.

[0112] S402. Obtain the matching results of person i and position j on the tth qualification factor respectively ,and ; The matching result between person i and position j on the sth ability factor , and the competency evaluation value of person i in position j ;

[0113] S403. Based on the above evaluation results, calculate the job matching degree of person i in job j :

[0114] ;

[0115] in, is a piecewise function and , which means that only when a person meets the basic job requirements, the ability and quality evaluation value and the comprehensive evaluation results based on the job competency model will be included in the MD calculation scope. Otherwise, the MD value is 0, which means that the person does not meet the basic job requirements and will not be included in the personnel allocation decision for the position. Q is the penalty coefficient, is the weight of the ability factor.

[0116] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. The human resource allocation system for construction enterprise project groups based on multi-objective optimization is characterized by: include: The model building module is used to build a multi-objective optimization model that includes maximizing the overall person-job matching of the project group, optimizing the age structure of the team members, and minimizing human resource costs, and setting allocation constraints for employees and positions; An information acquisition module, connected to the model building module by signals, for acquiring employee information and position information, and transmitting the acquired information to the model building module and the matching quantification module; A matching quantification module is connected to the information acquisition module and the model construction module by signal, and is used to receive the employee information and position information transmitted by the information acquisition module, quantify the person-position matching degree based on the position competency model and personnel portrait, and use the employee marginal contribution rate curve to evaluate the rationality of the team's personnel age structure, and transmit the quantification result signal to the model construction module and the solution generation module; a solution generation module, connected to the model construction module and the matching quantification module by signal, configured to receive the multi-objective optimization model transmitted by the model construction module and the quantitative result signal transmitted by the matching quantification module, solve the multi-objective optimization model using an improved NSGA-II algorithm, obtain a human resource allocation solution, and output the solution; The matching quantification module further includes an age structure assessment unit, which includes: The contribution rate curve fitting subunit is used to fit the employee marginal contribution rate curve based on the employee's historical performance data and age information; a contribution level determination subunit, configured to determine the expected contribution levels of employees of different age groups based on the employee marginal contribution rate curve; The team contribution calculation subunit is used to calculate the expected contribution of the team as a whole based on the age distribution of the current team members and the expected contribution level of employees in each age group; The structural rationality assessment sub-unit is used to compare the expected contribution of the team as a whole with the expected contribution under the preset reasonable age structure, and to evaluate the rationality of the current team age structure; The employee and position allocation constraints set by the model building module include: Each employee is assigned to at most one position, while some employees are allowed to remain unassigned; Each position must be assigned an employee; A priority weight coefficient is set for each project, where the priority weight coefficient is a value between 0 and 1, and the sum of the priority weight coefficients of all projects is 1.

2. The construction enterprise project group human resources allocation system based on multi-objective optimization according to claim 1 is characterized by: The employee information includes the employee's skill level, work experience, educational background, performance, professional qualification certificate and age information; the position information includes the position's job requirements, skill requirements, experience requirements, educational requirements, qualification certificate requirements and age structure requirements.

3. The construction enterprise project group human resources allocation system based on multi-objective optimization according to claim 2 is characterized in that: The matching quantification module includes: A competency model building unit, configured to build a competency model for each position based on the position information and determine the weight of each competency indicator; A personnel profile building unit, configured to build a personnel profile of the employee based on the employee information and quantify the scores of various competency indicators; A matching calculation unit is used to calculate the person-job matching score of each employee and each job based on the competency model of the job and the personnel portrait of the employee.

4. The construction enterprise project group human resources allocation system based on multi-objective optimization according to claim 1 is characterized in that: The method in which the solution generation module uses the improved NSGA-II algorithm to solve the multi-objective optimization model, obtains and outputs the human resource allocation solution includes: S101. Perform double-layer segmented encoding on chromosomes, where the chromosomes are used to simultaneously represent both personnel-project allocation information and personnel-position allocation information. The first layer of chromosomes represents the project to which the personnel are assigned, and the second layer of chromosomes represents the degree of match between the personnel and the position, expected salary, and age. S102. Use non-dominated sorting based on early rejection strategy to accelerate the sorting process of individuals in the population; S103. Performing individual selection based on an improved niche strategy, wherein the niche radius is dynamically determined based on the Euclidean distance between individuals in the population; S104. Perform a crossover operation using a partial matching crossover operator to ensure the genetic uniqueness of the chromosome after the crossover; perform a single-point mutation operation on the chromosome and perform a deduplication process to ensure the genetic uniqueness of the chromosome after the mutation; S105. Terminate the algorithm based on the number of iterations and the fitness value, and output the Pareto optimal solution set as the human resource allocation plan.

5. The construction enterprise project group human resources allocation system based on multi-objective optimization according to claim 1 is characterized in that: The objective function of the multi-objective optimization model constructed by the model construction module includes: S201. Maximize the overall project team's job matching , expressed as: ； in, Represents maximizing the overall person-job matching of the project group function, It represents the priority weight coefficient of project h, N is the total number of projects, n is the personnel number, and m is the position number; It represents the matching degree between position j of project h and person i; It is a 0-1 variable. When the value is 1, it means that person i has been assigned to position j. S202. Optimize the age structure of team members , expressed as: ； in, Represents the maximum age structure of the team members function, It represents the evaluation value of the rationality of the age structure of the team members of project h; S203. Minimize human resource costs , expressed as: ； in, Minimize human resource costs function, It represents the expected salary of person i for position j.

6. The construction enterprise project group human resources allocation system based on multi-objective optimization according to claim 4 is characterized in that: The method for individual selection based on the improved niche strategy comprises: S301. Calculate the Euclidean distance between individuals in a population: ； Where x and y represent two random individuals in the population. represents the Euclidean distance between x and y, xᵢ and yᵢ are the values ​​of the i-th objective function corresponding to individual x and individual y respectively; S302. After calculating the Euclidean distance between each individual and other individuals, the average of the minimum Euclidean distance values ​​of all individuals is taken as the initial value of the microhabitat radius; S303. Dynamically iterate and dynamically adjust the microhabitat radius. The microhabitat radius is gradually reduced according to the following formula: ; represents the current microhabitat radius, is a constant; gs is the current number of iterations, gs max is the maximum number of iterations; S304. When the Euclidean distance between two individuals x and y is less than the current microhabitat radius When , the fitness values ​​of the two individuals are compared, and the individual with lower fitness value is punished to reduce its probability of being selected.

7. The construction enterprise project group human resources allocation system based on multi-objective optimization according to claim 5 is characterized in that: The calculation process of the job matching degree includes the following sub-steps: S401. Analyze job attributes and determine the required employee capabilities. The set of positions requiring personnel in the project is denoted as J, and the set of available personnel for each position is denoted as I. S402. Obtain the matching results of person i and position j on the tth qualification factor respectively , the matching result between person i and position j on the sth ability factor , and the competency evaluation value of person i in position j ; S403. Based on the above evaluation results, calculate the job matching degree of person i in job j : ； in, is a piecewise function and ; Q is the penalty coefficient, is the weight of the ability factor.

Images