Comprehensive management system and method integrated with human resource outsourcing service

Abstract

The invention relates to the technical field of human resource management, in particular to a comprehensive management system and method for integrating human resource outsourcing services, and the system comprises a data integration and clustering mining module, a process optimization and cost control module, a dynamic employment risk assessment module, and a supplier cooperation and performance optimization module. The data integration and clustering mining module provides cost, efficiency and risk indexes through data integration and clustering mining; the process optimization and cost control module optimizes an outsourcing service process by using a genetic algorithm, sets a cost monitoring point, and realizes cost control and improvement of process efficiency; the dynamic employment risk assessment module constructs a Bayesian network based on the association rule analysis result, and dynamically assesses the employment risk; the supplier collaboration and performance optimization module establishes a platform, optimizes task allocation, evaluates supplier performance, and realizes efficient collaboration and survival of the suppliers. The method is used for solving the technical problem that the current outsourcing service management effect is poor.

Description

Technical Field

[0001] The present invention relates to the technical field of human resource management, and particularly to an integrated comprehensive management system and method for human resource outsourcing services. Background Art

[0002] The current outsourcing service management has the following problems: data related to outsourcing services may be scattered in various departments, lacking unified data integration and analysis; traditional outsourcing service management may rely on manual experience for process design and cost control, lacking scientific methods and tools to optimize processes and monitor costs in real time; existing risk assessment methods may be static and lagging, unable to dynamically assess employment risks based on real-time data and difficult to respond to potential risks in a timely manner. For this reason, the present invention proposes an integrated comprehensive management system and method for human resource outsourcing services. Summary of the Invention

[0003] The purpose of the present invention is to solve the problems in the background art, and to propose an integrated comprehensive management system and method for human resource outsourcing services.

[0004] To achieve the above purpose, the present invention adopts the following technical solutions:

[0005] An integrated comprehensive management system for human resource outsourcing services, comprising: a data integration and clustering mining module, a process optimization and cost control module, a dynamic employment risk assessment module, and a supplier collaboration and performance optimization module;

[0006] Data integration and clustering mining module: collect and integrate outsourcing service project data, generate cost indicators, efficiency indicators, and risk indicators, use clustering analysis algorithms to classify the outsourcing service project data by cost, efficiency, and risk, and perform relationship mining on the clustered outsourcing service project data through association rule mining algorithms;

[0007] Process optimization and cost control module: optimize the outsourcing service process through genetic algorithms, set fitness functions based on the clustering results of cost and efficiency indicators, iteratively search for the optimal process combination; set cost monitoring points and define cost monitoring functions, compare cost data in real time to determine and analyze whether it exceeds the expected range, and thus take measures;

[0008] Dynamic employment risk assessment module: construct a Bayesian network based on the association rule analysis results of outsourcing service project data, dynamically assess employment risks, and update the conditional probability distribution in real time to divide risk levels;

[0009] Supplier collaboration and performance optimization module: establish a supplier collaboration management platform, integrate supplier information, and use ant colony algorithms to optimize task allocation. At the same time, define supplier performance evaluation indicators, regularly evaluate supplier performance, and divide supplier cooperation levels.

[0010] Further, the data integration and clustering mining module collects and integrates the data of the outsourcing service projects, generates cost indicators, efficiency indicators and risk indicators, and classifies the cost, efficiency and risk of the outsourcing service project data by using the clustering analysis algorithm. The process of mining the relationships of the clustered outsourcing service project data through the association rule mining algorithm includes:

[0011] Collect the data of the outsourcing service projects, including the outsourcing service cost data set, the human resource information set, the outsourcing service provider data set and the external environment data set;

[0012] Obtain the collected data of the outsourcing service projects and label the data of the outsourcing service projects as osc;

[0013] Preprocess and integrate the data of the outsourcing service projects collected from different sources: extract relevant information from the preprocessed data of the outsourcing service projects and construct an index set of the outsourcing service projects, specifically including the cost indicator CI, the efficiency indicator EI and the risk indicator RI;

[0014] Apply the clustering analysis algorithm, and cluster the outsourcing service project data osc with the cost indicator, the efficiency indicator and the risk indicator as the clustering dimensions:

[0015] Based on the cost indicator, the efficiency indicator and the risk indicator, construct distance metric functions respectively, that is, the distance metric function D CI (osc 1 ,osc 2 ) of the cost indicator, the distance metric function D EI (osc 1 ,osc 2 ) of the efficiency indicator and the distance metric function D RI (osc 1 ,osc 2 ):

[0016]

[0017] In the formula, osc 1 ,osc 2 are the data of the outsourcing service project 1 and the data of the outsourcing service project 2 respectively, and υ is the index of the data of the outsourcing service project; the distance metric function of the cost indicator is used to measure the relative distance in cost between the data of the two outsourcing service projects 1 and the data of the outsourcing service project 2;

[0018]

[0019] In the formula, the distance metric function of the efficiency indicator is used to measure the relative distance in efficiency between the data of the two outsourcing service projects 1 and the data of the outsourcing service project 2;

[0020]

[0021] In the formula, the distance metric function of the risk index is used to measure the relative distance in risk between two pieces of outsourced service project data 1 and outsourced service project data 2;

[0022] Define the comprehensive distance metric function D 0 :

[0023] D 0 = ε 1 D CI (osc 1 , osc 2 ) + ε 2 D EI (osc 1 , osc 2 ) + ε 3 D RI (osc 1 , osc 2 ),

[0024] In the formula, ε 1 , ε 2 , ε 3 are the weight factors of the comprehensive distance metric function, and ε 1 + ε 2 + ε 3 = 1, which is determined according to the enterprise's emphasis on the cost index, efficiency index, and risk index of the outsourced service;

[0025] According to the comprehensive distance metric function D 0 cluster the outsourced service project data osc υ into the Z - class osc'υ = {osc' 1 , osc' 2 , …, osc' Z}, so that the project data within the same class are similar in terms of cost, efficiency, and risk; where Z is the number of clusters;

[0026] Within each cluster osc' υ , regarding each piece of outsourced service project data osc υ ∈ osc' υ as a transaction, then there is a transaction set In the formula, tc υ represents the υ - th transaction, and n υ is the number of projects in the cluster osc' υ ;

[0027] Within each cluster, further analyze the relationship between the outsourced service project data through the association rule mining algorithm:

[0028] Define the minimum support as ms and the minimum confidence as mc;

[0029] For the item set IS in the transaction set tc υ calculate the support sup(IS) of the item set:

[0030]

[0031] If sup(IS) ≥ ms, then IS is a frequent item set and is marked as IS';

[0032] For the frequent item set IS' and another item set IJ, and calculate the confidence conf(IS' → IJ):

[0033]

[0034] If conf(IS' → IJ) ≥ mc, then the association rule IS' → IJ is discovered.

[0035] Furthermore, the process optimization and cost control module optimizes the outsourcing service process through a genetic algorithm, sets the fitness function based on the clustering results of cost and efficiency indicators, and the process of iteratively finding the optimal process combination includes:

[0036] Determine the outsourcing service process links, decompose the current outsourcing service process into multiple adjustable links and operations, specifically including the job posting link, resume screening link, and interview arrangement link in the recruitment process, where each link contains different operation parameters, namely the selection of job posting channels and the criteria for resume screening;

[0037] Based on the clustering results corresponding to the cost indicators and efficiency indicators, determine the fitness function of the genetic algorithm; among them, the fitness function is defined as the weighted sum of the total time and total cost of the outsourcing service management process;

[0038] By continuously iterating the genetic algorithm, find the combination of outsourcing service process links and operation sequences that optimize the fitness function.

[0039] Furthermore, the process optimization and cost control module sets cost monitoring points and defines a cost monitoring function, and the process of comparing cost data in real time to determine and analyze whether it exceeds the expected range includes:

[0040] In the optimized outsourcing service process, set cost monitoring points for each outsourcing service process link, where the cost monitoring points determine the cost upper limit for each link in the outsourcing service process;

[0041] Define the cost monitoring function based on the set cost monitoring points;

[0042] By obtaining the output values of the cost monitoring functions corresponding to each cost monitoring point in real time and comparing them with the preset cost ceiling, it is determined whether each link in the outsourcing service process exceeds the expected cost range. If it exceeds the range, an alarm is triggered. At the same time, a cost control strategy is activated according to the alarm information.

[0043] Furthermore, the dynamic employment risk assessment module constructs a Bayesian network based on the analysis results of the association rules of the outsourcing service project data, dynamically assesses the employment risk, and updates the conditional probability distribution in real time. The process of dividing the risk levels includes:

[0044] Based on the analysis results of the association rules of the outsourcing service project data, determine the node relationships of the Bayesian network, including input nodes and output nodes;

[0045] In the Bayesian network, the conditional probability distribution between each node describes the dependence relationship between the nodes: for any given output node, randomly select two or more input nodes, and determine the joint probability distribution of these input nodes under the condition of the given output node, that is, obtain the joint probability distribution of the randomly selected input nodes when the output node is determined;

[0046] Through the Bayesian network, the conditional probability distribution between each node is updated in real time. For each Bayesian network output node, use the Bayesian formula to calculate its probability, so as to dynamically assess the probability and impact degree of the employment risk.

[0047] According to the employment risk assessment results, divide the risks into different levels, namely high, medium, and low levels. Screen out the high-risk levels and formulate corresponding countermeasures for the high-risk level situations. At the same time, connect the implementation of the risk countermeasures with the relevant links in the outsourcing service process.

[0048] Furthermore, the supplier collaboration and performance optimization module establishes a supplier collaboration management platform, integrates supplier information, and uses the ant colony algorithm to optimize the task allocation process, including:

[0049] Establish a supplier collaboration management platform, on which the information of each outsourcing service supplier is integrated, including the service capabilities of the supplier, historical performance data, and the current task allocation situation; at the same time, integrate the employment risk assessment results into the supplier collaboration management platform;

[0050] Use the ant colony algorithm to optimize the collaboration relationship between suppliers, that is, divide the outsourcing tasks into multiple subtasks and determine the attributes of each subtask;

[0051] According to the service capabilities and historical performance data of the suppliers, set the initial pheromone values for each supplier, where the pheromone value represents the attractiveness of the supplier when performing tasks;

[0052] Comprehensively evaluate the attractiveness of suppliers by establishing a function;

[0053] In the ant colony algorithm, ants represent different task assignment schemes. Ants move among suppliers, select the next supplier for task assignment according to the pheromone value and heuristic information, and update the pheromone value according to the task completion situation to find the optimal supplier task assignment scheme.

[0054] Furthermore, the process of the supplier collaboration and performance optimization module defining supplier performance evaluation indicators, regularly evaluating supplier performance, and dividing supplier cooperation levels includes:

[0055] Define supplier performance evaluation indicators, which cover the following aspects: cost control index, efficiency improvement index, and risk response index;

[0056] According to the set evaluation period, regularly collect supplier-related data, including cost control index, efficiency improvement index, and risk response index, and calculate and analyze the performance evaluation value of each supplier;

[0057] According to the calculated supplier performance evaluation value, divide suppliers into different cooperation levels, namely excellent, good, qualified, and unqualified.

[0058] An integrated human resource outsourcing service comprehensive management method, including:

[0059] Step 1: Collect and integrate outsourcing service project data, generate cost indicators, efficiency indicators, and risk indicators, use the clustering analysis algorithm to classify the outsourcing service project data for cost, efficiency, and risk, and use the association rule mining algorithm to mine the relationships of the clustered outsourcing service project data;

[0060] Step 2: Optimize the outsourcing service process through the genetic algorithm, set the fitness function based on the clustering results of cost and efficiency indicators, iteratively search for the optimal process combination; set cost monitoring points and define a cost monitoring function, compare the cost data in real time to judge and analyze whether it exceeds the expected range, and thus take measures;

[0061] Step 3: Construct a Bayesian network based on the association rule analysis results of outsourcing service project data, dynamically evaluate the employment risk, and update the conditional probability distribution in real time, divide the risk levels, and screen out the high-risk levels;

[0062] Step 4: Establish a supplier collaboration management platform, integrate supplier information, and use the ant colony algorithm to optimize task assignment. At the same time, define supplier performance evaluation indicators, regularly evaluate supplier performance, and divide supplier cooperation levels to achieve supplier collaboration and performance optimization.

[0063] Compared with the existing technologies, the advantages of the integrated human resource outsourcing service-based comprehensive management system and method provided by the present invention are as follows:

[0064] 1. By collecting and integrating the data of outsourcing service projects, the present invention generates cost indicators, efficiency indicators and risk indicators, classifies costs, efficiency and risks by using the clustering analysis algorithm, and conducts relationship mining through the association rule mining algorithm to reveal the internal connections and potential laws between the data, thereby improving the management efficiency;

[0065] 2. The present invention optimizes the outsourcing service process through the genetic algorithm, sets the fitness function, and iteratively searches for the optimal process combination; sets cost monitoring points and defines the cost monitoring function, and compares the cost data in real time to determine whether the analysis exceeds the expected range, realizing the automated and intelligent management of the process, reducing manual intervention, and improving work efficiency;

[0066] 3. By constructing a Bayesian network based on the results of the association rule analysis, the present invention dynamically evaluates the employment risk and divides the risk levels; establishes a supplier collaborative management platform, and optimizes the task allocation by using the ant colony algorithm. At the same time, it defines the supplier performance evaluation indicators, regularly evaluates the supplier performance, and divides the supplier cooperation levels, improving the service efficiency and quality while reducing the enterprise operation cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a module diagram of a comprehensive management system for integrated human resource outsourcing service proposed by the present invention.

[0068] Figure 2 It is a flowchart of a comprehensive management method for integrated human resource outsourcing service proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0069] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0070] Referring to Figure 1 , a comprehensive management system for integrated human resource outsourcing service, the system includes a data integration and clustering mining module, a process optimization and cost control module, a dynamic employment risk assessment module, and a supplier collaboration and performance optimization module;

[0071] Data Integration and Clustering Mining Module: Collect and integrate the data of outsourcing service projects, generate cost indicators, efficiency indicators and risk indicators, classify the cost, efficiency and risk of the outsourcing service project data by using the clustering analysis algorithm, and conduct relationship mining on the clustered outsourcing service project data through the association rule mining algorithm;

[0072] Process Optimization and Cost Control Module: Optimize the outsourcing service process through the genetic algorithm, set the fitness function based on the clustering results of cost and efficiency indicators, and iteratively search for the optimal process combination; Set cost monitoring points and define cost monitoring functions, compare cost data in real time to judge and analyze whether it exceeds the expected range, and then take measures;

[0073] Dynamic Employment Risk Assessment Module: Construct a Bayesian network based on the association rule analysis results of outsourcing service project data, dynamically assess employment risks, and update the conditional probability distribution in real time to divide risk levels;

[0074] Supplier Collaboration and Performance Optimization Module: Establish a supplier collaboration management platform, integrate supplier information, and use the ant colony algorithm to optimize task allocation to improve service efficiency. At the same time, define supplier performance evaluation indicators, regularly evaluate supplier performance, and divide supplier cooperation levels to achieve supplier collaboration and performance optimization.

[0075] The steps for the Data Integration and Clustering Mining Module to collect and integrate the data of outsourcing service projects, generate cost indicators, efficiency indicators and risk indicators, classify the cost, efficiency and risk of the outsourcing service project data by using the clustering analysis algorithm, and conduct relationship mining on the clustered outsourcing service project data through the association rule mining algorithm include:

[0076] Step 101: Collect the data of outsourcing service projects, including the outsourcing service cost data set, human resource information set, outsourcing service supplier data set and external environment data set;

[0077] In step 101, the service cost data set is a collection of cost details related to outsourcing services obtained from the finance department within the enterprise, covering outsourcing personnel salaries, benefits, training costs, recruitment costs, and the outsourcing budget allocation for different projects or departments; the human resources information set is a collection of basic information of outsourcing personnel (such as skill levels, job positions, working hours, etc.), personnel turnover rates, performance evaluation data, and internal demand forecasts and feedback information on outsourcing services provided by the human resources department; the outsourcing service provider data set is a collection of detailed records of outsourcing service delivery, such as the number of recruits, recruitment cycles, personnel arrival times, service response times, as well as the operating cost structure and human resources management strategies of the provider itself; the external environment data set is obtained from the labor law database for labor laws and regulations and their change records in various regions, and from industry research institutions for industry dynamic data, such as industry average salary levels, emerging skill requirements, and industry competition trends;

[0078] Step 102: Obtain the collected outsourcing service project data and label the outsourcing service project data as osc;

[0079] Step 103: Preprocess and integrate the outsourcing service project data collected from different sources: Extract relevant information from the preprocessed outsourcing service project data and construct an index set for the outsourcing service project, specifically including a cost index CI, an efficiency index EI, and a risk index RI;

[0080] In step 103, taking the cost index CI as an example, its calculation formula is:

[0081]

[0082] In the formula, is the preset weight coefficient of different cost indexes, ca is the actual total cost, ca 0 is the expected total cost, and Var(ca) is the cost variance, reflecting the stability of the cost;

[0083] Data preprocessing includes unified format conversion, handling missing values and error values; for example, unifying the date format to "YYYY-MM-DD", unifying the currency unit to RMB yuan, etc. For missing values, the method of filling with the mean (for numerical data) or the most common category (for categorical data) is used. For error values, they are corrected or deleted according to the logical relationship of the data;

[0084] Step 104: Use the clustering analysis algorithm, with the cost index, efficiency index, and risk index as the clustering dimensions, to cluster the outsourcing service project data osc (cluster the outsourcing service projects with similar costs, similar efficiencies, and facing the same risk types (such as legal risks, market risks, etc.) into one category);

[0085] In step 104, based on the cost metric, efficiency metric, and risk metric, distance metric functions are constructed respectively, namely, the distance metric function D of the cost metric CI (osc 1 , osc 2 ), the distance metric function D of the efficiency metric EI (osc 1 , osc 2 ), and the distance metric function D of the risk metric RI (osc 1 , osc 2 ):

[0086]

[0087] Wherein, osc 1 , osc 2 are the outsourcing service project data 1 and outsourcing service project data 2 respectively, and υ is the outsourcing service project data index; the distance metric function of the cost metric is used to measure the relative distance in cost between the two outsourcing service project data 1 and outsourcing service project data 2;

[0088]

[0089] Wherein, the distance metric function of the efficiency metric is used to measure the relative distance in efficiency between the two outsourcing service project data 1 and outsourcing service project data 2;

[0090]

[0091] Wherein, the distance metric function of the risk metric is used to measure the relative distance in risk between the two outsourcing service project data 1 and outsourcing service project data 2;

[0092] Define the comprehensive distance metric function D 0 :

[0093] D 0 = ε 1 D CI (osc 1 , osc 2 ) + ε 2 D EI (osc 1 , osc 2 ) + ε 3 D RI (osc 1 , osc 2 ),

[0094] Wherein, ε 1 , ε 2 , ε 3is the weight factor of the comprehensive distance metric function, and ε 1 + ε 2 + ε 3 = 1, which is determined according to the degree of importance of the enterprise for the outsourcing service cost index, efficiency index, and risk index;

[0095] According to the comprehensive distance metric function D 0 cluster the outsourcing service project data osc υ into Z classes osc'υ = {osc' 1 , osc' 2 , …, osc' Z}, so that the project data within the same class has similarity in terms of cost, efficiency, and risk; where Z is the number of clusters;

[0096] Within each cluster osc' υ , regard each outsourcing service project data osc υ ∈ osc' υ as a transaction, then there is a transaction set In the formula, tc υ represents the υ-th transaction, and n υ is the number of projects in the cluster osc' υ ;

[0097] Step 105, within each cluster, further analyze the relationship between the outsourcing service project data through the association rule mining algorithm, specifically:

[0098] Define the minimum support as ms and the minimum confidence as mc;

[0099] For the item set IS in the transaction set tc υ , calculate the support sup(IS) of the item set:

[0100]

[0101] If sup(IS) ≥ ms, then IS is a frequent item set and is marked as IS';

[0102] For the frequent item set IS' and another item set IJ, and calculate the confidence conf(IS' → IJ):

[0103]

[0104] If conf(IS'→IJ) ≥ mc, then the association rule IS'→IJ is discovered; for example, discovering the association rule between a specific combination of factors (such as regulations in a specific region combined with a specific business type and a specific supplier) and high costs or high risks, revealing the correlation rule between the efficiency of management process links and a specific data pattern (such as the distribution of the number of personnel and skill levels), and these mined relationships will provide a basis for subsequent optimization decisions.

[0105] The process optimization and cost control module optimizes the outsourcing service process through a genetic algorithm, sets the fitness function based on the clustering results of cost and efficiency indicators, and iteratively searches for the optimal process combination; setting cost monitoring points and defining a cost monitoring function, and the steps of comparing cost data in real time to determine whether it exceeds the expected range and thus taking measures include:

[0106] Step 201: Determine the outsourcing service process links, decompose the current outsourcing service process into multiple adjustable links and operations, specifically including the job posting link, resume screening link, and interview arrangement link in the recruitment process, where each link contains different operation parameters, namely the selection of job posting channels and the criteria for resume screening;

[0107] Step 202: Based on the clustering results corresponding to the cost indicator and the efficiency indicator, determine the fitness function of the genetic algorithm; among them, the fitness function is defined as the weighted sum of the total time and total cost of the outsourcing service management process;

[0108] Step 203: Continuously iterate by running the genetic algorithm to find the combination of outsourcing service process links and operation sequences that optimize the fitness function;

[0109] In steps 201 - 203, set the outsourcing service process to include g links (corresponding to each process link), and mark the outsourcing service process link as SP j , where g is the number of outsourcing service process links, j is the index of the outsourcing service process link, and j = 1, 2,..., g;

[0110] Each outsourcing service process link SP j contains several operations SO jl , where l is the operation index in the outsourcing service process link SP j , and l = 1, 2,..., n j , n j is the number of operations in the outsourcing service process link SP j ;

[0111] Define the time consumption of each operation SO jl as ot jl , and the cost consumption as oc jl , then the outsourcing service process link SPj Total time AT j : Total cost AC j :

[0112] The fitness function SF is defined as:

[0113]

[0114] where w AC and w AT are the weights of cost and efficiency in the outsourcing service of the enterprise respectively, and w AC + w AT = 1;

[0115] For each operation SO jl , there are op jl adjustable parameters, and the adjustable parameters are denoted as oa jlh , where h = 1, 2,..., op jl , h is the index of the adjustable parameters existing in the operation SO jl , and op jl is the number of adjustable parameters existing in the operation SO jl ; for example, in the job posting operation SO 11 in the recruitment process, assuming it is the first operation in the recruitment process link SP 1 , then the selection parameter oa 111 of the job posting channel (different channels are represented by different values), the resume screening operation SO 12 , the standard parameter oa 121 of resume screening, etc.;

[0116] After determining the fitness function SF, genetic algorithm iteration is carried out. In each round of iteration, the fitness function value SF d of each individual (i.e., the process operation combination and sequence) is calculated, where d is the iteration index of the genetic algorithm;

[0117] Set the individual GX d = (gx d1 , gx d2 ,..., gx dq ) to represent the operation parameter combination of the d-th individual, then

[0118]

[0119] where gx d represents the operation parameter value of the d-th individual, q is the total number of parameters, and

[0120] For example, during the optimization of the recruitment process, after multiple rounds of iteration, it was found that changing the resume screening standard parameter from education priority (parameter set to ) to skill priority (parameter set to ) could shorten the recruitment cycle (i.e., decrease), and reduce the recruitment cost (i.e., decrease); ultimately, the optimal combination and sequence of process operations that maximize the fitness function SF were found;

[0121] Step 204: In the optimized outsourcing service process, set cost monitoring points for each link of the outsourcing service process, where the cost monitoring points define the cost ceiling for each link in the outsourcing service process;

[0122] Step 205: Define a cost monitoring function based on the set cost monitoring points;

[0123] Step 206: By obtaining the output value of the cost monitoring function corresponding to each cost monitoring point in real time and comparing it with the preset cost ceiling, determine whether each link in the outsourcing service process exceeds the expected cost range. If it exceeds the range, trigger an alarm. At the same time, start a cost control strategy based on the alarm information;

[0124] In steps 204 - 206, in the optimized outsourcing service process, for each operation SO jl , set cost monitoring points for each link of the outsourcing service process; for example, in the job posting operation SO 11 in the recruitment process, if the recruitment website channel is selected (parameter set to ), then the preset cost ceiling is If the social media channel is selected (parameter set to ), then the preset cost ceiling is etc.;

[0125] Define the cost monitoring function for operation SO jl at time t as AC jl (t), where the cost monitoring function AC jl (t) is a function of all relevant cost factors in operation SO jl , and AC jl (t) = Γ(ot jl , oa jl , t), where Γ is the function mapping relationship, which depends on the specific cost composition and operation parameters;

[0126] Obtain the output value of the cost monitoring function AC jl corresponding to each monitoring point for operation SO jl (t) in real time, and compare it with the preset cost ceiling AC' jlCompare. If AC jl (t) > AC' jl , then trigger the early warning mechanism; among them, the early warning function WF jl (t) of the early warning mechanism is:

[0127]

[0128] When WF jl (t) = 1, it means that the early warning is triggered;

[0129] According to the early warning information, start the cost control strategy; for example, if the cost of the recruitment channel is too high (i.e., AC 11 (t) > AC' 11 ), then define the adjustment coefficient related to market data and enterprise needs, and according to the market data and enterprise needs, switch to other recruitment channels with higher cost performance; if in the interview arrangement link, the cost of the operation is too high (i.e., ), then set the adjustment coefficient to and adjust the number of interview rounds or the interview method (such as changing from on-site interview to video interview);

[0130] Obtain the adjusted cost monitoring function

[0131]

[0132] At the same time, the adjusted number of interview rounds In the formula, is rounding down.

[0133] The dynamic employment risk assessment module constructs a Bayesian network based on the association rule analysis results of the outsourcing service project data, dynamically assesses the employment risk, and updates the conditional probability distribution in real time. The steps of dividing the risk level include:

[0134] Step 301: Based on the association rule analysis results of the outsourcing service project data, determine the node relationship of the Bayesian network, including input nodes and output nodes. For example, take regulatory changes, market fluctuations, supplier stability, outsourcing personnel skill matching degree, etc. as input nodes, and take employment risks (such as legal dispute risks, personnel shortage risks, performance non-compliance risks, etc.) as output nodes, where the employment risk is obtained according to the risk indicators;

[0135] Step 302. In the Bayesian network, the conditional probability distribution between nodes describes the dependency relationship between nodes: For any given output node, randomly select two or more input nodes, and determine the joint probability distribution of these input nodes under the condition of the given output node, that is, obtain the joint probability distribution of the randomly selected input nodes when the output node is determined.

[0136] Among them, the obtained joint probability distribution provides information about the co-occurrence probability of input nodes under the condition of the given output node. For example, when the regulation changes to a certain type (such as strengthened labor rights protection) and the market fluctuation is positive (increased market demand), what is the probability of reduced supplier stability, and further what is the probability of increased risk of personnel shortage.

[0137] Step 303. With the continuous input of new data (i.e., real-time regulation updates, market dynamic monitoring data, the latest operation data of suppliers, etc.), the conditional probability distribution between nodes in the Bayesian network is updated in real time. For each output node of the Bayesian network, use the Bayesian formula to calculate its probability, so as to dynamically evaluate the probability and impact degree of employment risk.

[0138] Step 304. According to the employment risk assessment results, classify the risks into different levels, namely high, medium, and low levels. Screen out the high-risk level and formulate corresponding countermeasures for the high-risk level situation. At the same time, connect the implementation of the risk countermeasures with relevant links in the outsourcing service process. For example, the review of contract terms can be achieved through smart contract technology, and the adjustment of the recruitment plan can be fed back to the recruitment process optimization link in the outsourcing service process optimization and cost control module to ensure the coordinated operation of the entire outsourcing service management system.

[0139] In Steps 301 - 304, define as the set of input nodes of the Bayesian network, then the corresponding set of output nodes of the Bayesian network is

[0140] In the formula, can represent regulation changes, market fluctuations, supplier stability, outsourcing personnel skill matching degree, etc., and n 1 is the number of input nodes of the Bayesian network, can represent legal dispute risk, personnel shortage risk, performance non-compliance risk, etc., and n 2 is the number of output nodes of the Bayesian network.

[0141] According to the clustering results of the outsourcing service project data, establish the relationship between the input nodes and output nodes of the Bayesian network: For each output node of the Bayesian network, assume it is related to the input node There is a functional relationship between where the functional relationship represents the probability that, given an input node , the output node occurs. J 1 , J 2 are the indices of the output node and input node of the Bayesian network, respectively;

[0142] For example, for the legal dispute risk by 1 , when the regulation change bx 1 is of a certain type (such as strengthened labor rights protection) and the market fluctuation bx 2 is positive (increased market demand), then it is necessary to determine the probability that the supplier stability bx 3 decreases, that is

[0143] BP(bx 3 = decrease | bx 1 = strengthened labor rights protection, bx 2 = positive), and then determine the probability that the risk of personnel shortage by 2 increases, that is

[0144] BP(by 2 = increase | bx 1 = strengthened labor rights protection, bx 2 = positive, bx 3 = decrease);

[0145] Analyze the conditional probability distribution between the nodes in the Bayesian network: Extract three different random input nodes and label them as bx 1 , bx 2 , bx 3 to represent factors such as regulation change, market fluctuation, supplier stability, and outsourcing personnel skill matching degree, where these factors are considered potential causes or preconditions affecting employment risk;

[0146] Define BP(bx 1 , bx 2 , bx 3 | by 2 ) to represent the joint probability distribution of the input nodes bx 2 , bx 1 , bx 2 , bx 3 given the output node by

[0147] For example, when the regulation change is of a certain type (such as strengthened labor rights protection) and the market fluctuation is positive (increased market demand), then what is the probability that the supplier stability decreases, and further what is the probability that the risk of personnel shortage increases,

[0148] Let A be {bx 1 = enhanced labor protection rights and interests, bx 2 = positive},

[0149] B be {bx 3 = decreased},

[0150] E be {by 2 = increased},

[0151] then determine BP(BA) and BP(EA,B);

[0152] As new data is continuously input (i.e., real-time regulatory updates, market dynamic monitoring data, the latest operation data of suppliers), the Bayesian network updates the conditional probability distribution between nodes in real-time; mark the new data as {bd 1 ,bd 2 ,bd 3}, then the updated conditional probability distribution is BP(bx 1 ,bx 2 ,bx 3 |by 2 ,bd 1 ,bd 2 ,bd 3 );

[0153] Dynamically evaluate the probability and impact degree of employment risks: For each output node of the Bayesian network calculate its probability using Bayes' formula, where Bayes' formula is:

[0154]

[0155] Establish a risk level function

[0156] Map the employment risk assessment results to different levels, namely high, medium, and low levels:

[0157] Set level thresholds LT 1 ,LT 2 , and LT 1 < LT 2 ;

[0158] Compare and analyze the probability value calculated by Bayes' formula with the level thresholds: If then is low level; if then is medium level; if then is high level;

[0159] Response Strategies for High - Risk Levels If the legal dispute risk is assessed as high, the response strategies include immediately reviewing the terms of the outsourcing service contract to ensure that all terms comply with the latest regulatory requirements; if the personnel shortage risk is high, the response strategy is to adjust the recruitment plan for outsourced personnel, expand the recruitment scope or increase the salary and benefits to attract more talents;

[0160] The implementation of risk response strategies is connected to relevant links in the outsourcing service process, and response strategies are set up with connection functions for the outsourcing service process For example, the review of contract terms can be achieved through smart contract technology, and the adjustment of the recruitment plan can be fed back to the recruitment process optimization link in the outsourcing service process optimization and cost control module.

[0161] The steps for the supplier collaboration and performance optimization module to establish a supplier collaboration management platform, integrate supplier information, optimize task allocation using the ant colony algorithm, define supplier performance evaluation indicators, regularly evaluate supplier performance, and classify supplier cooperation levels are as follows:

[0162] Step 401: Establish a supplier collaboration management platform, which integrates information of each outsourcing service supplier, including the supplier's service capabilities, historical performance data, and current task allocation situation; among them, the supplier's service capabilities: refer to the historical cooperation records between the supplier and the platform or other enterprises to evaluate its performance and service capabilities in actual projects; historical performance data: collect feedback and evaluations from end - customers on the services provided by the supplier as a supplement to historical performance data; current task allocation situation: update and display the task allocation situation in real - time on the supplier collaboration management platform to ensure that all relevant parties can obtain the latest information in a timely manner; at the same time, integrate the employment risk assessment results into the supplier collaboration management platform as an important reference basis for supplier selection and task allocation;

[0163] Step 402: Use the ant colony algorithm to optimize the collaboration relationship between suppliers, that is, divide the outsourcing tasks into multiple subtasks and determine the attributes of each subtask;

[0164] Step 403: Set an initial pheromone value for each supplier according to the supplier's service capabilities and historical performance data, where the pheromone value represents the attractiveness of the supplier when performing tasks;

[0165] Step 404: Comprehensively evaluate the attractiveness of suppliers by establishing a function;

[0166] Step 405: In the ant colony algorithm, ants represent different task allocation schemes. Ants move among suppliers, select the next supplier for task allocation based on pheromone values and heuristic information, and update the pheromone values according to the task completion situation to find the optimal supplier task allocation scheme, so that the overall service efficiency reaches the best state;

[0167] Among them, the labor risk assessment result is used as part of the heuristic information to guide ants to more carefully consider risk factors when selecting the next supplier for task allocation; heuristic information is a key factor in the ant colony algorithm for task allocation in the supplier collaboration management platform. It is based on multiple aspects such as the service capabilities of suppliers, historical performance data, the matching degree between subtask attributes and suppliers, and risk assessment results;

[0168] In steps 401 - 405, obtain the outsourcing task TA and divide it into n 3 sub - tasks ta, that is where n 3 is the number of sub - tasks in the outsourcing task TA;

[0169] For each sub - task ta i , define its attribute vector In the formula, i is the sub - task index, is the dimension of the sub - task attribute vector. For example, a i1 is the task difficulty (represented by a numerical value from 1 - 5, 1 being the easiest and 5 being the most difficult), a i2 is the required skill (which can be represented by different encodings for different skills), a i3 is the estimated completion time;

[0170] Suppose there are k suppliers, denoted as S = {S u |S 1 , S 2 , …, S k}, in the formula, u is the supplier index and k is the number of suppliers;

[0171] For each supplier S u , its basic information is I u , and the service capacity vector is ι is the dimension of the service capacity vector. For example, c u1 is the proficiency in a certain specific skill, c u2 is the available manpower scale;

[0172] Obtain the historical performance data P u , the current task allocation situation L u ;

[0173] Set an initial pheromone value τ for each supplier S according to the service capabilities and historical performance data of the suppliers u ; u0 ;

[0174] Establish a function f(P u , C u ) to comprehensively evaluate the attractiveness of the suppliers, then τ u0 = f(P u , C u );

[0175] Define that ant m represents different task assignment schemes, where m = 1, 2, …, M, and M is the total number of ants;

[0176] In each iteration r (r is the iteration index), ant m starts from a random supplier S u0 ;

[0177] When ant m is at supplier S u , the probability ξ u' that it selects the next supplier S uu' is:

[0178]

[0179] In the formula, α and β are parameters that adjust the relative importance of pheromone and heuristic information respectively, τ u' (r) is the pheromone value of the next supplier S u' at the r-th iteration, η uu' (r) is the heuristic information (for example In the formula, ψ 1uu' represents the distance between supplier S u and the next supplier S u' , ψ 2uu' represents the cost of assigning tasks from supplier S u to the next supplier S u' , o is the index of the neighboring suppliers of supplier S u , and N u is the set of neighboring suppliers of supplier S u (i.e., the set of suppliers that can be selected to assign tasks);

[0180] As the number of iterations increases, the pheromone value is updated according to the task completion situation. Define the completion situation Q and the completion threshold Q 0 ;

[0181] Set Δτ uu' as the amount of pheromone deposited by ant m when moving from supplier S u to the next supplier S u'The amount of pheromone left on the path. If the task assignment is successful and the completion is good, i.e., Q > Q 0 , then

[0182]

[0183] where path m is the path taken by ant m;

[0184] The pheromone update formula is:

[0185]

[0186] where ρ is the pheromone evaporation coefficient, and 0 < ρ < 1;

[0187] By continuous iteration, gradually find the optimal supplier task assignment scheme to make the overall service efficiency reach the best state;

[0188] Step 406: Define supplier performance evaluation indicators. The supplier performance evaluation indicators cover the following aspects: cost control index, efficiency improvement index, and risk response index. Among them, the cost control index includes the deviation rate of the actual cost from the budget cost. The efficiency improvement index is measured by data such as the performance evaluation results of outsourced personnel and the results of customer satisfaction surveys. The risk response index can be evaluated according to the response speed of the supplier when facing risks and the effectiveness of the response measures;

[0189] It can be understood that the actual cost is obtained through the enterprise financial management system, which records the actual expenditure of each cost item. The budget cost is estimated through demand analysis for the required human resources, materials, equipment, outsourcing services, etc. The response speed and the effectiveness of the response measures of the supplier when facing risks are obtained through the evaluation of the supplier's risk response ability by a third-party agency. The performance evaluation results of outsourced personnel are obtained through the enterprise's negotiation with the outsourcing service provider to establish a regular assessment mechanism to evaluate the performance of outsourced personnel. The results of customer satisfaction surveys can be obtained by using online evaluation platforms (such as e-commerce platforms, social media, etc.) to collect customer evaluations. These platforms usually have a large number of user evaluations and data, which can be used as an important reference for customer satisfaction surveys;

[0190] Step 407: Regularly collect supplier-related data according to the set evaluation period, including the cost control index, efficiency improvement index, and risk response index, and calculate and analyze the performance evaluation value of each supplier;

[0191] Step 408: According to the calculated supplier performance evaluation values, suppliers are divided into different cooperation levels, namely, excellent, good, qualified, and unqualified. For excellent suppliers, more business shares, more favorable cooperation conditions (such as early payment, lower management fees, etc.) are given, and more resources are tilted on the collaborative management platform (such as priority assignment of tasks, more training opportunities, etc.); for good suppliers, normal cooperation relationships are maintained and appropriate incentives are given; for qualified suppliers, they are monitored and required to improve; for unqualified suppliers, assistance is provided (such as providing improvement suggestions, organizing training, etc.), and if the performance does not improve within a certain period of time, the cooperation is reduced or terminated, so as to continuously optimize the outsourcing service management, reduce risks and improve overall efficiency;

[0192] In steps 406-408, based on the clustering results of the cost index, the efficiency index and the risk index, the cost control index K is defined respectively. 1 , efficiency improvement index K 2 and risk response index K 3 ; Among them, the cost control index Efficiency Improvement Index In the formula, jx u is the performance evaluation result of the outsourced personnel, sd u For customer satisfaction survey results, is the weight coefficient; risk response index Where sv u For supplier S u The speed of response when facing risks, me u For supplier S u The effectiveness of the response measures, is the weight coefficient;

[0193] Joint Cost Control Index K 1 , efficiency improvement index K 2 and risk response index K 3 Get Supplier S u The performance evaluation value F(S u ):

[0194]

[0195] In the formula, They are cost control index K 1 , efficiency improvement index K 2 and risk response index K 3 The weight factor of

[0196] According to the set evaluation period (such as quarterly, annual), regularly collect relevant data of suppliers and calculate the performance evaluation value F(S u );

[0197] Based on the performance evaluation value F(S u ), divide suppliers into different cooperation levels: set the threshold for the excellent level as F ex , the threshold for the good level as F go , the threshold for the qualified level as F qu , and the threshold for the unqualified level as F un ;

[0198] If F(S u ) ≥ F ex , then the supplier S u is of the excellent level, and more business shares and more preferential cooperation conditions (such as early payment, lower management fees, etc.) are given, and more resource inclinations are given on the collaborative management platform (such as preferential task allocation, more training opportunities, etc.); if F go ≤ F(S u ) < F ex , then the supplier S u is of the good level, maintain a normal cooperation relationship and give appropriate incentive measures; if F qu ≤ F(S u ) < F go , then the supplier S u is of the qualified level, monitor it and require improvement; if F(S u ) < F qu , then the supplier S u is of the unqualified level, provide assistance (such as providing improvement suggestions, organizing training, etc.), and if the performance does not improve within a certain period, reduce cooperation or terminate the cooperation relationship.

[0199] Refer to Figure 2 , a comprehensive management method integrating human resource outsourcing services, including:

[0200] Step 1: Collect and integrate outsourcing service project data, generate cost indicators, efficiency indicators and risk indicators, classify the outsourcing service project data by cost, efficiency and risk using the clustering analysis algorithm, and mine the relationships of the clustered outsourcing service project data through the association rule mining algorithm;

[0201] Step 2: Optimize the outsourcing service process through the genetic algorithm, set the fitness function based on the clustering results of the cost and efficiency indicators, and iteratively search for the optimal process combination; set cost monitoring points and define the cost monitoring function, compare the cost data in real time to judge and analyze whether it exceeds the expected range, and thus take measures;

[0202] Step 3: Construct a Bayesian network based on the association rule analysis results of the outsourcing service project data, dynamically evaluate the employment risk, and update the conditional probability distribution in real time, classify the risk levels, and screen out the high-risk levels.

[0203] Step 4: Establish a supplier collaborative management platform, integrate supplier information, and optimize task allocation using the ant colony algorithm. At the same time, define supplier performance evaluation indicators, regularly evaluate supplier performance, and classify supplier cooperation levels to achieve supplier collaboration and performance optimization.

[0204] In addition, the formulas involved above are all calculated by removing the dimension and taking their numerical values. It is a formula obtained by software simulation of a large amount of collected data to be closest to the real situation. The proportionality coefficient in the formula and each preset threshold in the analysis process are set by those skilled in the art according to the actual situation or obtained by simulation of a large amount of data; the size of the proportionality coefficient is a specific numerical value obtained by quantifying each parameter for subsequent comparison. Regarding the size of the proportionality coefficient, it depends on the amount of sample data and the processing coefficients initially set by those skilled in the art for each group of sample data; as long as it does not affect the proportional relationship between the parameters and the quantified values.

[0205] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the differences between each embodiment and other embodiments are emphasized. In particular, for the device embodiments, since they are basically based on the method embodiments, they are described relatively simply, and the relevant parts can be referred to the corresponding descriptions in the method embodiments.

[0206] For the convenience of description, the above device is described by dividing it into various units according to functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0207] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can be implemented in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can be implemented in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0208] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0209] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0210] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operational steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0211] Secondly, in the attached drawings of the disclosed embodiments of the present invention, only the structures related to the disclosed embodiments are involved. For other structures, reference can be made to the general design. Without conflict, the same and different embodiments of the present invention can be combined with each other;

[0212] Finally, the above are only the preferred specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and all should be covered by the protection scope of the present invention.

Claims

1. A comprehensive management system integrating human resource outsourcing services, characterized by: It includes data integration and cluster mining module, process optimization and cost control module, dynamic employment risk assessment module and supplier collaboration and performance optimization module; Data integration and cluster mining module: collect and integrate outsourcing service project data, generate cost indicators, efficiency indicators and risk indicators, use cluster analysis algorithms to classify outsourcing service project data in terms of cost, efficiency and risk, and use association rule mining algorithms to perform relationship mining on clustered outsourcing service project data; Process optimization and cost control module: optimize outsourcing service processes through genetic algorithms, set fitness functions based on clustering results of cost and efficiency indicators, and iteratively search for the optimal process combination; Set cost monitoring points and define cost monitoring functions, compare cost data in real time to determine whether the analysis exceeds the expected range, and take measures accordingly; Dynamic employment risk assessment module: Based on the association rule analysis results of outsourced service project data, a Bayesian network is constructed to dynamically assess employment risks, and the conditional probability distribution is updated in real time to classify risk levels; Supplier collaboration and performance optimization module: Establish a supplier collaboration management platform, integrate supplier information, and use ant colony algorithm to optimize task allocation. At the same time, define supplier performance evaluation indicators, regularly evaluate supplier performance, and divide supplier cooperation levels.

2. According to claim 1, a comprehensive management system for integrated human resource outsourcing services is characterized by: The data integration and cluster mining module collects and integrates outsourcing service project data, generates cost indicators, efficiency indicators and risk indicators, uses cluster analysis algorithms to classify outsourcing service project data in terms of cost, efficiency and risk, and uses association rule mining algorithms to perform relationship mining on clustered outsourcing service project data. The process includes: Collect outsourcing service project data, including outsourcing service cost data set, human resource information data set, outsourcing service provider data set and external environment data set; Obtain the collected outsourcing service project data, and mark the outsourcing service project data as osc; Preprocess and integrate the outsourcing service project data collected from different sources: extract information from the preprocessed outsourcing service project data and construct an indicator set for the outsourcing service project, including cost indicator CI, efficiency indicator EI and risk indicator RI; Using cluster analysis algorithm, the outsourcing service project data osc is clustered with cost index, efficiency index and risk index as clustering dimensions: Based on the cost index, efficiency index and risk index, distance measurement functions are constructed respectively, that is, the distance measurement function D of the cost index CI (osc1, osc2), distance measurement function D of efficiency index EI (osc1, osc2) and the distance measurement function D of the risk indicator RI (osc1,osc2), as follows: In the formula, osc1 and osc2 are outsourcing service project data 1 and outsourcing service project data 2 respectively, and υ is the outsourcing service project data index; the distance measurement function of the cost index is used to measure the relative distance between the two outsourcing service project data 1 and outsourcing service project data 2 in terms of cost; In the formula, the distance measurement function of the efficiency index is used to measure the relative distance between the two outsourcing service project data 1 and outsourcing service project data 2 in terms of efficiency; In the formula, the distance measurement function of the risk index is used to measure the relative distance between the two outsourcing service project data 1 and the outsourcing service project data 2 in terms of risk; Define the comprehensive distance metric function D0: D0=ε1D CI (osc1,osc2)+ε2D EI (osc1,osc2)+ε3D RI (osc1,osc2), Where ε1, ε2, ε3 are weight factors of the comprehensive distance metric function, and ε1+ε2+ε3=1; According to the comprehensive distance measurement function D0, the outsourcing service project data osc υ Clustered into Z types osc'υ={osc'1,osc'2,…,osc' Z }, so that the project data in the same category have similarities in cost, efficiency and risk; where Z is the number of clusters; In each cluster osc' υ In the osc υ ∈osc' υ As a transaction, there is a transaction set In the formula, tc υ represents the υth transaction, n υ For clustering osc' υ The number of items in Within each cluster, the relationship between the outsourced service project data is further analyzed through the association rule mining algorithm: Define the minimum support as ms and the minimum confidence as mc; For transaction set tc υ Item set IS in , calculate the support of the item set sup(IS): If sup(IS)≥ms, IS is a frequent item set and is marked as IS'; For a frequent itemset IS' and another itemset IJ, and Calculate the confidence conf(IS'→IJ): If conf(IS'→IJ)≥mc, the association rule IS'→IJ is found.

3. According to claim 1, the integrated management system for integrated human resource outsourcing services is characterized by: The process optimization and cost control module optimizes the outsourcing service process through genetic algorithms, sets the fitness function based on the clustering results of cost and efficiency indicators, and iterates the process of finding the optimal process combination, including: Determine the outsourcing service process links and break down the current outsourcing service process into multiple adjustable links and operations, including the job posting link, resume screening link, and interview arrangement link in the recruitment process. Each link contains different operating parameters, i.e. the selection of job posting channels and the criteria for resume screening; Based on the clustering results corresponding to the cost index and the efficiency index, the fitness function of the genetic algorithm is determined; wherein the fitness function is defined as the weighted sum of the total time and the total cost of the outsourcing service management process; By running the genetic algorithm for continuous iteration, the combination of outsourcing service process links and operation sequence that optimizes the fitness function is found.

4. The integrated management system for integrated human resource outsourcing services according to claim 3 is characterized by: The process optimization and cost control module sets cost monitoring points and defines cost monitoring functions. The process of comparing cost data in real time to determine whether the analysis exceeds the expected range includes: In the optimized outsourcing service process, a cost monitoring point is set for each link of the outsourcing service process, wherein the cost monitoring point determines the cost upper limit for each link in the outsourcing service process; Define cost monitoring functions based on set cost monitoring points; By obtaining the output value of the cost monitoring function corresponding to each cost monitoring point in real time and comparing it with the preset cost upper limit, it is determined whether each link in the outsourced service process exceeds the expected cost range. If it exceeds the range, an early warning is triggered. At the same time, the cost control strategy is initiated based on the early warning information.

5. The integrated management system for integrated human resource outsourcing services according to claim 2, characterized in that: The dynamic employment risk assessment module constructs a Bayesian network based on the association rule analysis results of the outsourced service project data, dynamically assesses employment risks, and updates the conditional probability distribution in real time. The process of dividing the risk level includes: Based on the association rule analysis results of the outsourced service project data, determine the node relationship of the Bayesian network, including input nodes and output nodes; In a Bayesian network, the conditional probability distribution between nodes describes the dependency between nodes: for any given output node, two or more input nodes are randomly selected, and the joint probability distribution of these input nodes under the given output node condition is determined, that is, the joint probability distribution of the randomly selected input nodes is obtained when the output node is determined; The conditional probability distribution between nodes is updated in real time through the Bayesian network. For each Bayesian network output node, the probability is calculated using the Bayesian formula, thereby dynamically evaluating the probability of employment risk. According to the employment risk assessment results, the risks are divided into different levels, namely high, medium and low. The high-risk levels are screened out and corresponding response strategies are formulated for the high-risk levels. At the same time, the implementation of the risk response strategy is linked to the relevant links in the outsourcing service process.

6. The integrated management system for integrated human resource outsourcing services according to claim 5, characterized in that: The supplier collaboration and performance optimization module establishes a supplier collaboration management platform, integrates supplier information, and uses the ant colony algorithm to optimize task allocation. The process includes: Establish a supplier collaborative management platform that integrates the information of each outsourcing service provider, including the supplier's service capabilities, historical performance data, and current task allocation; at the same time, integrate the employment risk assessment results into the supplier collaborative management platform; Use ant colony algorithm to optimize the collaboration between suppliers, that is, divide the outsourced task into multiple subtasks and determine the attributes of each subtask; According to the supplier's service capabilities and historical performance data, an initial pheromone value is set for each supplier, where the pheromone value represents the supplier's attractiveness when performing tasks; Comprehensively evaluate the attractiveness of suppliers by establishing functions; In the ant colony algorithm, ants represent different task allocation schemes. Ants wander among suppliers, select the next supplier for task allocation based on pheromone values ​​and heuristic information, and update the pheromone value based on the task completion status to find the optimal supplier task allocation scheme. Among them, heuristic information is the key factor used by the ant colony algorithm to guide ants to select the next supplier in task allocation on the supplier collaborative management platform.

7. The integrated management system for integrated human resource outsourcing services according to claim 1, characterized in that: The supplier collaboration and performance optimization module defines supplier performance evaluation indicators, regularly evaluates supplier performance, and classifies supplier cooperation levels. The process includes: Define supplier performance evaluation indicators, which cover the following aspects: cost control index, efficiency improvement index and risk response index; According to the set evaluation cycle, regularly collect supplier-related data, including cost control index, efficiency improvement index and risk response index, and calculate and analyze the performance evaluation value of each supplier; Based on the calculated supplier performance evaluation values, suppliers are divided into different cooperation levels, namely excellent, good, qualified, and unqualified.

8. A comprehensive management method for integrated human resource outsourcing services, characterized in that: The method applied to a comprehensive management system integrating human resource outsourcing services as claimed in any one of claims 1 to 7 comprises: Step 1: Collect and integrate outsourcing service project data, generate cost indicators, efficiency indicators and risk indicators, use clustering analysis algorithms to classify outsourcing service project data by cost, efficiency and risk, and use association rule mining algorithms to perform relationship mining on the clustered outsourcing service project data; Step 2: Optimize the outsourcing service process through genetic algorithms, set the fitness function based on the clustering results of cost and efficiency indicators, and iterate to find the optimal process combination; set cost monitoring points and define cost monitoring functions, compare cost data in real time to determine whether the analysis exceeds the expected range, and then take measures; Step 3: Based on the association rule analysis results of the outsourced service project data, a Bayesian network is constructed to dynamically evaluate employment risks, and the conditional probability distribution is updated in real time to classify risk levels; Step 4: Establish a supplier collaborative management platform, integrate supplier information, and use the ant colony algorithm to optimize task allocation. At the same time, define supplier performance evaluation indicators, regularly evaluate supplier performance, and divide supplier cooperation levels.

Images