Hybrid office strategy optimization system and method based on Bayesian decision model

Abstract

The invention relates to a hybrid office strategy optimization system based on a Bayesian decision model, and the system comprises a data collection module which collects the basic historical information of employees and the real-time use condition of an office; the Bayesian decision engine obtains the prior probability of the station utilization rate in different office modes according to the collected basic historical information of the employees and calculates a likelihood function, and the likelihood function is used for calculating the influence of different office modes on the production efficiency according to the real-time use condition of the office space, calculating the marginal likelihood and calculating the working efficiency; according to the prior probability, the likelihood function and the marginal likelihood, working efficiency posterior probabilities in different office modes are calculated; and the decision optimization module is used for determining and recommending the optimal office mode of the employees and the optimal number of office days per week based on the posterior probability output by the Bayesian decision engine. According to the invention, through intelligent decision support and a flexible adjustment mechanism, synchronous improvement of working efficiency, office space utilization rate and staff work-family life balance is realized.

Description

Technical Field

[0001] The present invention relates to the field of intelligent management technology, and in particular to a hybrid office strategy optimization system and method based on a Bayesian decision model. Background Art

[0002] More and more companies are adopting a hybrid office model, where employees flexibly switch between office and remote work. While this model improves work flexibility, it also brings new management challenges, such as how to optimize office space utilization, how to balance employees' work-life needs, and how to ensure work efficiency under different office modes. However, most existing office management systems rely on fixed rules and manual adjustments, making it difficult to provide dynamic optimization decisions based on changes in actual needs. In addition, the lack of a real-time response mechanism to employee feedback results in the inability to adjust office modes and space configurations in a timely manner, which in turn affects work efficiency and employee satisfaction. Therefore, there is an urgent need for a system based on intelligent decision-making that can dynamically optimize office modes and space configurations in a hybrid office environment to cope with changing work needs and environmental challenges. Summary of the invention

[0003] The present invention provides a hybrid office strategy optimization system and method based on a Bayesian decision model to solve how to dynamically optimize office modes and space configurations in a hybrid office environment based on an intelligent decision-making system.

[0004] The present invention provides a hybrid office strategy optimization system based on a Bayesian decision model, comprising: a data collection module, which collects basic historical information of employees and real-time usage of office space; a Bayesian decision engine, which obtains the prior probability of workstation usage under different office modes according to the collected basic historical information of employees, calculates a likelihood function, wherein the likelihood function calculates the influence of different office modes on production efficiency according to the real-time usage of the office space, calculates marginal likelihood, wherein the marginal likelihood is the weighted efficiency level under all possible office modes, and calculates the posterior probability of work efficiency under different office modes according to the prior probability, the likelihood function and the marginal likelihood; and a decision optimization module, which determines and recommends the optimal office mode and the optimal office days per week of employees based on the posterior probability output by the Bayesian decision engine.

[0005] Furthermore, the basic historical information of the employees includes: an employee information database, a work nature assessment module, and an employee office mode preference collection module. The employee information database includes the basic information, job positions, and historical performance appraisal results of the employees. The work nature assessment module includes whether teamwork is required and whether reliance on equipment or resources is required.

[0006] Furthermore, the data collection module includes an office space utilization monitoring unit to monitor the real-time utilization of the office space, including workstation occupancy rate and meeting utilization rate.

[0007] Furthermore, the posterior probability of work efficiency under different office modes is:

[0008] in,

[0009] P(D i |E i , Office Mode j ) is the likelihood function, data D i Given office mode j and efficiency E i The probability of

[0010] P(D i |E i , Office Mode j ) is the prior probability, the work efficiency of employee i in office mode j;

[0011] P(D i ) is the marginal likelihood,

[0012] Furthermore, when the employee's performance rating follows a normal distribution, its mean is μ j , with standard deviation σ j , the likelihood function is:

[0013]

[0014] Further, based on the posterior probability output by the Bayesian decision engine, the optimal working mode and the optimal working days per week of the employees are determined and recommended: an objective function is established, Among them, E i is the work efficiency of employee i, W i is the weight of employee i, S j is the utilization rate of office space j, U j is the priority of space usage, α and β are weight coefficients, and the weight coefficients are confirmed according to the posterior probability.

[0015] Furthermore, the hybrid office strategy optimization system based on the Bayesian decision model includes: a visualization interface for displaying a management view and an employee view, wherein the management view includes a data visualization display of office mode allocation, office space utilization, and employee performance trends, and the employee view includes an office mode, work arrangement, and work feedback unit recommended by the system for employees to view.

[0016] The present invention also provides a hybrid office strategy optimization method based on a Bayesian decision model, comprising:

[0017] Collect employees' basic historical information and real-time usage of office space; obtain a priori probability of workstation usage under different office modes based on the collected basic historical information of employees, calculate likelihood function, the likelihood function calculates the impact of different office modes on production efficiency based on the real-time usage of the office space, calculate marginal likelihood, the marginal likelihood is the weighted efficiency level under all possible office modes, calculate posterior probability of work efficiency under different office modes based on the prior probability, the likelihood function and the marginal likelihood; determine and recommend the optimal office mode and optimal office days per week for employees based on the posterior probability output by the Bayesian decision engine.

[0018] Furthermore, the posterior probability of work efficiency under different office modes is in,

[0019] P(D i |E i , Office Mode j ) is the likelihood function, data D i Given office mode j and efficiency E i The probability of

[0020] P(E i |Office Mode j ) is the prior probability, the work efficiency of employee i in office mode j;

[0021] P(D i ) is the marginal likelihood,

[0022] Furthermore, when the employee's performance rating follows a normal distribution, its mean is μ j , with standard deviation σ j , the likelihood function is:

[0023]

[0024] Further, based on the posterior probability output by the Bayesian decision engine, the optimal working mode and the optimal working days per week of the employees are determined and recommended: an objective function is established, Among them, E i is the work efficiency of employee i, W i is the weight of employee i, S j is the utilization rate of office space j, U j is the priority of space usage, α and β are weight coefficients, and the weight coefficients are confirmed according to the posterior probability.

[0025] The hybrid office strategy optimization system and method based on the Bayesian decision model provided by the present invention formulates the optimal hybrid office strategy for the enterprise through data collection, analysis and dynamic decision-making, so as to achieve the best combination of employee productivity and work-life balance, while maximizing the utilization efficiency of office space. The system can automatically assign the most suitable office mode according to the employee's job position, job nature, historical work performance and his / her preference for office mode, and optimize the office days of hybrid office personnel. DETAILED DESCRIPTION

[0026] The following will combine the scheme to clearly and completely describe the technical scheme in this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work belong to the scope of protection of this application.

[0027] The present invention provides a hybrid office strategy optimization system based on a Bayesian decision model, comprising: a data collection module, which collects basic historical information of employees and real-time usage of office space; a Bayesian decision engine, which obtains the prior probability of workstation usage under different office modes according to the collected basic historical information of employees, calculates a likelihood function, wherein the likelihood function calculates the influence of different office modes on production efficiency according to the real-time usage of the office space, calculates marginal likelihood, wherein the marginal likelihood is the weighted efficiency level under all possible office modes, and calculates the posterior probability of work efficiency under different office modes according to the prior probability, the likelihood function and the marginal likelihood; and a decision optimization module, which determines and recommends the optimal office mode and the optimal office days per week of employees based on the posterior probability output by the Bayesian decision engine.

[0028] The data collection module is the foundation of the system and is used to collect multi-dimensional data related to employee office mode decisions. The collected basic historical information of employees includes: employee information database, job nature evaluation module, and employee office mode preference collection module. The employee information database includes employees' basic information, job positions, and historical performance appraisal results. These data are used to set prior probabilities to help the Bayesian decision engine evaluate the impact of different office modes on employee productivity.

[0029] Table 1 is the content of the employee information database provided by the embodiment of the present invention. Table 1

[0030]

[0031]

[0032] Employee information database data collection methods: When an employee joins the company, basic information of the employee (such as name, department, position, etc.) is collected and entered into the system through the enterprise human resources management system (HRMS). Employee performance data is collected through regular performance evaluation or assessment processes and recorded in the HRMS. Employee office preference data is collected through regular employee questionnaires or one-on-one interviews. The questionnaires can be conducted through online survey tools, and the interview data is recorded by the human resources department. The IT department determines the employee's proficiency in remote office technology through the employee's daily work performance or specific skills assessment (such as post-training tests) and records it in the system.

[0033] The basic information of employees is collected once when they join the company, and any subsequent changes (such as job transfers, department changes) need to be updated in real time. Performance appraisal data is collected quarterly or semi-annually to reflect the latest work performance of employees. Surveys on office mode preferences are conducted every six months or before and after major company policy adjustments (such as the introduction of new office models) to ensure the accuracy and timeliness of the data. The initial assessment is conducted on the proficiency of remote office technology when the employee takes up the post, and then re-evaluated annually or every two years as needed (such as technology updates, job adjustments).

[0034] The work nature assessment module is used to assess the nature of work in different positions, including whether teamwork is required, whether it depends on specific equipment or resources, etc. These data help determine which office model is more suitable for a certain position.

[0035] Table 2 is a work property evaluation module provided by an embodiment of the present invention

[0036] Table 2

[0037]

[0038] Collect information about the nature of the job through in-depth interviews or seminars with department managers or job leaders. Historical data analysis (such as employee task allocation and team collaboration tool usage) can be used to supplement and verify the assessment results. Conduct a comprehensive analysis once a year, especially when the company adjusts job responsibilities or introduces new positions, and update them immediately.

[0039] Through questionnaire surveys, we ask employees about their teamwork, equipment reliance, and other details in their daily work. We conduct this survey every six months to understand employees' latest feelings and changes in their job duties.

[0040] The employee preference collection module for office mode collects employee preferences for office mode through questionnaires or direct communication, such as preference for working from home, flexible schedule, etc. This preference information will be used together with other data for decision-making to ensure that the allocation of office mode can take into account the personal wishes of employees.

[0041] Table 3 shows the content of the employee preference collection module for office mode provided in an embodiment of the present invention.

[0042] Table 3

[0043]

[0044] Use online survey tools to regularly send questionnaires to employees to investigate their preferences for work patterns, commute times, work-family life balance, etc. Conduct this survey every six months, or additionally when the company introduces new work policies or adjusts existing work patterns.

[0045] For key positions or senior employees, one-on-one interviews are conducted through HR or direct supervisors to gain a deeper understanding of their office preferences and reasons. Interview records must be updated to the system in a timely manner. This should be conducted once a year or when there is a major change in an employee's position (such as promotion or transfer).

[0046] The data collection module includes an office space utilization monitoring unit, which monitors the use of office space in real time, including workstation occupancy rate, conference room utilization rate, etc. These data are used to optimize the utilization of office space and avoid waste or shortage of resources.

[0047] Table 4 is the monitoring content of the office space utilization monitoring unit provided by the embodiment of the present invention.

[0048]

[0049] Install occupancy sensors (such as infrared, ultrasonic or thermal sensors) in office spaces to monitor the real-time use of workstations and meeting rooms. Sensor data is automatically recorded and transmitted to the system. Sensor data is collected in real time and can be updated to the system hourly or daily as needed.

[0050] Access control data: Records the time and frequency of employee access to office areas through employee access cards or biometric systems. This data can be combined with occupancy sensor data to ensure the accuracy of office space utilization data. Collected in real time and automatically synced to the system every hour or day.

[0051] Manual Recording: In the absence of sensors or smart devices, the occupancy of workstations and meeting rooms can be manually recorded by office administrators. Manually recorded data will be entered into the system regularly. Recording can be done at fixed time periods daily or weekly, and manually entered into the system after recording is completed, and summarized and analyzed weekly or monthly.

[0052] The Bayesian decision engine is the core of the system, which uses Bayesian decision theory and algorithms to optimize the hybrid office strategy. The Bayesian decision engine includes three main modules: prior probability setting, likelihood function calculation, and posterior probability update.

[0053] Using historical data and expert evaluation, we assign initial probabilities to the work efficiency under each office mode. Assuming there are three office modes (entirely in the office, hybrid office, and fully remote office), we can set the prior probability of each mode as a preliminary estimate. i )(Work efficiency of employee i in office mode j)

[0054] P(E i |Office Modej)=Initial Estimateij

[0055] Assume that in historical data, the average productivity of full office work is 80%, that of mixed office work is 75%, and that of full remote work is 70%. Then P(E i |FullyOffice)=0.80,P(E i |Hybrid)=0.75,P(E i |Remote)=0.70.

[0056] Use statistical methods to calculate the probability of employee performance data under a given office mode. Common likelihood function forms include normal distribution, Bernoulli distribution, etc., depending on the data type.

[0057] Set the likelihood function P(D i |E i ,OfficeMode j )(Data D i Given office mode j and efficiency E i probability): P(D i |E i , Office Mode j )=Likelihood Functionij

[0058] Assume that the employee's performance rating follows a normal distribution with mean μ j , with standard deviation σ j , calculate the likelihood function:

[0059]

[0060] If the employee's actual rating in full office mode is 85 points, and the mean and standard deviation of the office mode are 80 points and 5 points respectively, then:

[0061]

[0062] When employees work completely in the office, the probability of observing data with a score of 85 is 3.7%. The likelihood function calculation under different office modes compares the effects of different office modes to determine which mode is more suitable for employees. In an embodiment of the present invention, the likelihood function is 0.037 in the full office mode, 0.15 in the remote office mode, and 0.25 in the hybrid office mode. In this case, the hybrid office mode is more suitable.

[0063] Based on the newly collected data, the probability distribution of employees' work efficiency under different office modes is updated, and the posterior probability is calculated by combining the prior probability and the likelihood function using Bayes' theorem:

[0064]

[0065] Where: P(D i |E i , Office Mode j ) is the likelihood function; P(E i |Office Mode j ) is the prior probability;

[0066] P(D i ) is the marginal likelihood, which can be calculated by weighting all possible efficiency levels:

[0067]

[0068] The calculated posterior probability is used to adjust the employee's office mode allocation. Suppose we want to update the probability of employees' work efficiency in the full office mode. i |Fully Office)=0.80, the likelihood function value is 0.037, and the marginal likelihood P(D i )≈0.25:

[0069]

[0070] This posterior probability reflects the updated value of employee productivity in full office mode based on the new data.

[0071] The core task of the decision optimization module is to formulate and optimize hybrid office strategies based on the output of the Bayesian decision engine to balance the work efficiency of employees and the utilization of office space. This module includes two sub-modules: strategy optimization and space utilization optimization. The strategy optimization sub-module ensures that different positions and employees can maximize their work efficiency under different office modes by formulating hybrid office strategies. Generate recommended office mode strategies based on the posterior probability output of the Bayesian decision engine. Consider the work efficiency of each employee under different office modes and generate the optimal office mode allocation plan. Set office mode priorities for different positions. For example, positions with high collaboration needs give priority to working in the office, while positions with high autonomy give priority to remote work. Evaluate the effects of different office mode strategies and use simulated data for prediction to ensure that the strategies can achieve their goals.

[0072] Establish an objective function to maximize the weighted sum of overall work efficiency and employee satisfaction.

[0073]

[0074] Among them, E i is the work efficiency of employee i, W i is the weight of employee i, S j is the utilization rate of office space j, U j is the priority of space usage, and α and β are weight coefficients.

[0075] In the embodiment of the present invention, the company has three positions (A, B, C). After the Bayesian decision engine analysis, position A has a higher remote working efficiency, position B has the best hybrid working effect, and position C is suitable for full office work. The strategy optimization submodule will generate a recommendation plan based on this information, such as: Position A: mainly remote work, occasionally return to the office for meetings; Position B: hybrid working mode, working in the office three days a week; Position C: completely working in the office.

[0076] Optimize the utilization of office space to ensure the maximum use of office resources and avoid space waste or shortage. Forecast future office space needs based on employees' office mode allocation and actual usage. Consider employees' work schedules and optimize the use of workstations and meeting rooms; reasonably configure office space based on demand forecast results, and optimize the layout and number of workstations and meeting rooms. Monitor the real-time use of office space and adjust space configuration to cope with dynamic changes.

[0077] Space utilization calculation: Calculate space utilization based on actual occupancy data of workstations and meeting rooms, and use time series analysis or regression models to predict future space requirements.

[0078] In the embodiment of the present invention, the company has 100 workstations and 10 meeting rooms. Existing data shows that 20% of the workstations are not used every day. The space utilization optimization submodule adjusts the configuration plan to reduce unused space through real-time data monitoring and predictive analysis. For example, if it is predicted that the peak workstation utilization rate will increase to 80% in the next month, it may be necessary to reconfigure the workstations and increase the number of workstations in high-demand areas.

[0079] The Bayesian decision model is used to analyze the collected data, update the weights of various conditional probabilities and decision factors, and generate office mode recommendations for different employees or teams. By analyzing data trends and fluctuations, potential low work efficiency or unreasonable space use can be detected, and corresponding warnings or adjustment suggestions can be triggered.

[0080] The system provides two main views: management view and employee view. Each view displays relevant information and functions according to different user needs to support decision-making and operations.

[0081] The dashboard in the management view uses line charts or bar charts to show the work efficiency trends under different office modes, and uses pie charts or bar charts to show the actual use of workstations and meeting rooms. Display employee names, departments, current office modes, and recommended modes in a table. Visual charts show the distribution ratio of office modes in different departments or positions, using bar charts or stacked bar charts. Allow management to manually adjust employees' office modes and view the effects of the adjustments.

[0082] Provides real-time office space usage and future demand forecasts to help optimize space configuration and resource allocation. Real-time occupancy: Displays the real-time usage of current workstations and conference rooms, using heat maps or real-time data tables. Uses line charts or forecast charts to display future space demand forecasts, including demand for workstations and conference rooms; provides space configuration suggestions based on forecast data, and displays recommended space adjustment plans.

[0083] The management view shows key indicators such as the overall performance of the current office model, space utilization, and employee satisfaction. The system also generates future trend forecasts and optimization suggestions based on the analysis results to help management make strategic decisions. Generate a detailed office model performance report covering the office model performance of each department and team to support management in conducting comprehensive business analysis.

[0084] The employee view provides employees with personalized office mode information and related functions, supporting self-management and feedback. It displays personal work arrangements and satisfaction surveys to help employees understand and adjust their own work modes.

[0085] Display employees' current office mode arrangements, including office hours and locations, provide system-recommended adjustment suggestions, and allow employees to apply for changes; provide a calendar view of personal work arrangements, display office modes and arrangements, allow employees to view future work arrangements, and make necessary adjustments; display employees' work arrangements, including specific dates for in-office, hybrid office, and remote office; allow employees to apply for adjustments to personal work arrangements and view the status of the application.

[0086] Display employee satisfaction scores on the current office model, including work efficiency, work-life balance, etc., and provide employees with the option to submit detailed feedback and improvement suggestions; provide notifications and reminders about office model adjustments, satisfaction surveys, and other related information to ensure that employees are aware of and respond to important updates in the system in a timely manner. Display notifications and reminders sent by the system, including office model changes, satisfaction surveys, etc.; allow employees to set preferences for notifications and reminders, such as email notifications or in-app reminders.

[0087] Through the employee view, employees can view their office mode arrangements, workstation allocation, personal performance and satisfaction feedback, etc. The system will also provide corresponding resources and support to help employees better adapt to the hybrid office mode. Employees can submit feedback directly through the interface, and the system will automatically adjust office arrangements or generate improvement suggestions based on the feedback.

[0088] Embodiment 1

[0089] The Bayesian formula is used to calculate the posterior probability of different office modes (such as full office, hybrid office, and remote office), and the optimal workstation configuration is adjusted in combination with the objective function. The calculation process of the Bayesian formula is as follows:

[0090] Among them: P(Hi|D) is the posterior probability of the i-th office mode (such as full office work, hybrid office, remote office) under the new data D; P(D|Hi) is the possibility of the occurrence of data D (such as actual workstation utilization rate) when the i-th office mode is assumed to be established, which is called the likelihood function; P(Hi) is the prior probability of the office mode, that is, the possibility of workstation utilization rate under each office mode obtained based on historical data; P(D) is the marginal probability of the occurrence of data D, which is used to normalize the Bayesian formula.

[0091] In the first embodiment of the present invention, for the full office working mode (assuming that the current working mode is mainly full office working), the prior probability P(Office) = 0.7 (that is, historically 70% of the time employees work in the office), the prior probability of the hybrid office mode P(Hybrid) = 0.2, and the prior probability of the full remote office mode P(Remote) = 0.1.

[0092] Calculate the likelihood function P(D|Hi): In the office working mode, the workstation utilization rate is 80%, then P(Data|Office) may be higher, for example, 0.9; in the hybrid working mode, the workstation utilization rate is 60%, then P(Data|Hybrid) may be 0.6; in the remote working mode, the workstation utilization rate is 40%, then P(Data|Remote) is 0.4.

[0093] Calculate the posterior probability: Assuming that the observed data D is that the workstation utilization rate increases to 80%, calculate the posterior probability using the Bayesian formula to determine the likelihood of each office mode under the new data.

[0094]

[0095] The posterior probability of each office mode is calculated, which will provide a basis for subsequent decision-making.

[0096] The objective function is used to measure the optimal solution for space configuration. Common objective functions can be maximizing space utilization or minimizing the number of unused workstations. We can use Bayesian posterior probability to weigh different workstation configuration solutions and choose the one with the highest expected utility.

[0097] The objective function can be the expected value of the workstation utilization rate:

[0098] E(U)=P(Office|Data)·U(Office)+P(Hybrid|Data)·U(Hybrid)+P(Remote|Data)·U(Remote)

[0099] Where: E(U) is the expected value of space utilization, P(Office|Data), P(Hybrid|Data), and P(Remote|Data) are the posterior probabilities under different office modes, and U(Office), U(Hybrid), and U(Remote) are the workstation utilization rates under different office modes, respectively.

[0100] In the first embodiment of the present invention, the current predicted workstation utilization rate in the next month is 80% during the peak period. The posterior probability obtained according to the Bayesian formula is: P(Office|Data)=0.6 (the probability of full office work is 60%); P(Hybrid|Data)=0.3 (the probability of a hybrid office mode is 30%); P(Remote|Data)=0.1 (the probability of remote work is 10%).

[0101] Workstation utilization rate: U(Office) = 0.9 (the utilization rate of workstations in office working mode is 90%); U(Hybrid)

[0102] =0.7 (the utilization rate of workstations in the hybrid office mode is 70%); U(Remote) = 0.5 (the utilization rate of workstations in the remote office mode is 50%).

[0103] The expected station utilization rate is:

[0104] E(U)=0.6·0.9+0.3·0.7+0.1·0.5=0.54+0.21+0.05=0.80

[0105] Based on this expected utilization rate, decide how to adjust the number of workstations. If the predicted future utilization rate is 80%, the number of workstations needs to be adjusted to: the number of workstations required = 100*0.80 = 80 workstations.

[0106] Companies can adjust the number or distribution of workstations in the office, for example, by moving workstations from less-used areas to high-demand areas. Some unused workstations can be reduced and reconfigured into shared workstations or flexible workstations for more employees to use during peak hours.

[0107] Based on the posterior probability and the objective function, the enterprise can make the following decisions: Based on the posterior probability and expected utilization, configure 80 workstations for full office work, and set up shared office areas in flexible areas for employees in a hybrid office mode. If the posterior probability of the hybrid office mode is high, the enterprise can configure flexible shared workstations to meet the needs of some employees. For employees working remotely, reduce unnecessary fixed workstations, thereby improving the overall utilization of workstations.

[0108] The present invention realizes the efficient operation of the hybrid office strategy optimization system based on the Bayesian decision model through detailed implementation steps and processes. The system forms a closed-loop management from data collection, analysis, decision generation, feedback and adjustment to visual display, ensuring the dynamic optimization and continuous improvement of office strategies, and ultimately achieving the simultaneous improvement of corporate work efficiency and employee satisfaction.

Claims

1. A hybrid office strategy optimization system based on a Bayesian decision model, characterized in that: include: Data collection module, which collects basic historical information of employees and real-time usage of office space; The Bayesian decision engine obtains the prior probability of the workstation utilization rate under different office modes based on the basic historical information of the collected employees, calculates the likelihood function, the likelihood function calculates the impact of different office modes on production efficiency based on the real-time usage of the office space, calculates the marginal likelihood, the marginal likelihood is the weighted efficiency level under all possible office modes, and calculates the posterior probability of work efficiency under different office modes based on the prior probability, the likelihood function and the marginal likelihood; The decision optimization module determines and recommends the optimal working mode and the optimal working days per week for employees based on the posterior probability output by the Bayesian decision engine.

2. The hybrid office strategy optimization system based on the Bayesian decision model according to claim 1, characterized in that: The basic historical information of the employees includes: an employee information database, a work nature assessment module, and an employee office mode preference collection module. The employee information database includes the basic information, job position, and historical performance appraisal results of the employees. The work nature assessment module includes whether teamwork is required and whether reliance on equipment or resources is required.

3. The hybrid office strategy optimization system based on the Bayesian decision model according to claim 1, characterized in that: The data collection module includes an office space utilization monitoring unit, which monitors the real-time usage of office space, including workstation occupancy rate and meeting utilization rate.

4. The hybrid office strategy optimization system based on the Bayesian decision model according to claim 1, characterized in that: The posterior probability of work efficiency under different office modes is: in, P(D i |E i , Office Mode j ) is the likelihood function, data D i Given office mode j and efficiency E i probability; P(E i |Office Mode j ) is the prior probability, the work efficiency of employee i in office mode j; P(D i ) is the marginal likelihood, 5. The hybrid office strategy optimization system based on the Bayesian decision model according to claim 4, characterized in that: When the employee's performance rating follows a normal distribution, its mean is μ j , with standard deviation σ j , the likelihood function is:

6. The hybrid office strategy optimization system based on the Bayesian decision model according to claim 1, characterized in that: Based on the posterior probability output by the Bayesian decision engine, determine and recommend the optimal working mode and optimal working days per week for employees: establish an objective function, Among them, E i is the work efficiency of employee i, W i is the weight of employee i, S j is the utilization rate of office space j, U j is the priority of space usage, α and β are weight coefficients, and the weight coefficients are confirmed according to the posterior probability.

7. The hybrid office strategy optimization system based on the Bayesian decision model according to claim 1, characterized in that: Also includes: A visualization interface is used to display a management view and an employee view. The management view includes a data visualization of office mode allocation, office space utilization, and employee performance trends. The employee view includes a unit for employees to view the office mode, work schedule, and work feedback recommended by the system.

8. A hybrid office strategy optimization method based on a Bayesian decision model, characterized in that: include: Collect basic historical information of employees and real-time usage of office space; According to the basic historical information of employees collected, the prior probability of the utilization rate of workstations under different office modes is obtained, and the likelihood function is calculated. The likelihood function calculates the impact of different office modes on production efficiency according to the real-time utilization of the office space, and the marginal likelihood is calculated. The marginal likelihood is the weighted efficiency level under all possible office modes, and the posterior probability of work efficiency under different office modes is calculated according to the prior probability, the likelihood function and the marginal likelihood; Based on the posterior probability output by the Bayesian decision engine, the optimal working mode and optimal working days per week for employees are determined and recommended.

9. The hybrid office strategy optimization method based on the Bayesian decision model according to claim 8, characterized in that: The posterior probability of work efficiency under different office modes is: in, P(D i |E i , Office Mode j ) is the likelihood function, data D i Given office mode j and efficiency E i probability; P(D i |E i , Office Mode j ) is the prior probability, the work efficiency of employee i in office mode j; P(D i ) is the marginal likelihood, 10. The hybrid office strategy optimization method based on the Bayesian decision model according to claim 9, characterized in that: When the employee's performance rating follows a normal distribution, its mean is μ j , with standard deviation σ j , the likelihood function is:

11. The hybrid office strategy optimization method based on the Bayesian decision model according to claim 9, characterized in that: Based on the posterior probability output by the Bayesian decision engine, determine and recommend the optimal working mode and optimal working days per week for employees: establish an objective function, Among them, E i is the work efficiency of employee i, W i is the weight of employee i, S j is the utilization rate of office space j, U j is the priority of space usage, α and β are weight coefficients, and the weight coefficients are confirmed according to the posterior probability.

Images