A universal multifunctional decision-making system and method

Abstract

The present invention relates to the field of intelligent decision-making technology, and discloses a universal multifunctional decision-making system and method, comprising an information sharing subsystem and a decision-making subsystem; the information sharing subsystem comprises a personal information sharing module and an inter-group information sharing module; the decision-making subsystem comprises a task collection module and a task decision module; the task collection module is used to collect tasks to be assigned; the task decision module comprises a task analysis unit, a personnel analysis unit, and a matching unit; the task analysis unit is used to divide the tasks to be assigned and obtain multiple subtasks in a time sequence; the personnel analysis unit is used to determine the employee's task acceptance time zone and task acceptance type; the matching unit is used to match subtasks with employees in a time sequence and generate task allocation suggestions. The present invention can intelligently set and optimize the enterprise task allocation plan based on the actual situation of employees, and is rich in functions and highly versatile, which helps to achieve synergy and efficiency in enterprise work.

Description

Technical Field

[0001] The present invention relates to the field of intelligent decision-making technology, and in particular to a universal multifunctional decision-making system and method. Background Art

[0002] Today, digitalization is accelerating businesses into an era of high-frequency competition. Exponential technological advancements, accelerating industrial transformation, and a relentless wave of global disruption will continuously alter people's lives and the trajectory of business development, accelerating a new wave of digital transformation—a reality requiring every business to undergo digital transformation.

[0003] However, the digitalization process for enterprises still faces challenges that urgently need to be addressed. While most companies have introduced various digital management systems, such as OA systems, PM systems, CRM systems, and ERP systems, these systems primarily manage procedural tasks according to the company's own rules or centralize online data management. They are unable to leverage digital analysis to optimize management decisions, nor can they provide intelligent decision-making recommendations for optimizing enterprise management or improving employee experience.

[0004] For example, when assigning tasks within a company, tasks are often assigned by management or requested by employees themselves. However, subjective decision-making and cognitive biases (such as miscalculation of employee workload and inaccurate statistics of employee skills) can lead to irrational task allocation, impacting team efficiency. Furthermore, when multiple departments are under management, many employees' skills are limited by information barriers between departments and thus go unutilized, making it difficult to find suitable employees for certain tasks in a timely manner. In particular, in the context of big data, conventional approaches that focus on individual employees are no longer reliable when making task allocation decisions. Summary of the Invention

[0005] The present invention aims to provide a universal multifunctional decision-making system and method, which can intelligently set and optimize the enterprise task allocation plan based on the actual situation of employees, and has rich functions and strong versatility, which helps to achieve synergy and efficiency in enterprise work.

[0006] To achieve the above objectives, the present invention provides the following basic solutions.

[0007] Option 1

[0008] A general-purpose multifunctional decision-making system includes an information sharing subsystem and a decision-making subsystem; the information sharing subsystem includes a personal information sharing module and an inter-group information sharing module; the personal information sharing module is used to automatically collect employees' personal task information and allow employees to upload their personal task information; the inter-group information sharing module is used to collect personnel arrangement information, task arrangement information, and historical project information from various departments and groups in the enterprise;

[0009] The decision-making subsystem includes a task collection module and a task decision module; the task collection module is used to collect tasks to be assigned; the task decision module includes a task analysis unit, a personnel analysis unit and a matching unit; the task analysis unit is used to divide the tasks to be assigned according to time sequence and section, and obtain multiple subtasks in time sequence; the personnel analysis unit is used to determine the employee's task acceptance time zone and task acceptance type based on personal task information and data collected by the inter-group information sharing module; the matching unit is used to match subtasks with employees in time sequence according to the analysis results of the task analysis unit and the personnel analysis unit, and generate task allocation suggestions.

[0010] Furthermore, the personal task information includes the employee's current task data and completed task data; the current task data includes the planned completion time of the task; and the personnel analysis unit is further used to quantify the task carrying capacity of each employee based on the employee's completed task data.

[0011] Furthermore, when matching subtasks with employees in chronological order, the following steps are included: based on the timing of the subtasks, first match them with the employee's task acceptance time zone, and filter to obtain an initial matching pair; then, based on the initial matching pair, match them with the employee's task acceptance type, and filter to obtain a target matching pair.

[0012] Furthermore, the initial matching pair is a many-to-many matching pair, and the target matching pair is a one-to-many matching pair; after the target matching pair is screened, the task carrying capacity occupied by the subtask is also evaluated, and employees with more task carrying capacity margin are selected to match the subtask.

[0013] Furthermore, when quantifying the task carrying capacity of each employee and evaluating the task carrying capacity occupied by subtasks, a preset neural network model is used for quantitative evaluation; the preset neural network model is provided with an input layer, a self-attention network layer, an RNN network layer, a first fully connected layer, a Relu function layer, a second fully connected layer and an output layer.

[0014] Furthermore, when quantifying the task carrying capacity of each employee, the preset neural network model takes as input the employee's completed task data and outputs the threshold value of the employee's task carrying capacity; when evaluating the task carrying capacity occupied by a subtask, the preset neural network model takes as input the employee's completed task data and subtask data, and outputs the proportion of the subtask in the employee's task carrying capacity.

[0015] Furthermore, the task analysis unit adopts a preset classification model to divide the tasks to be assigned; the preset classification model is a CNN-RNN joint model; and before dividing the tasks to be assigned, an initial judgment is also made on the tasks to be assigned; the initial judgment includes: searching for target words in the data of the tasks to be assigned, and if the search result shows that the number of target word matches is less than 2, then directly treating the task to be assigned as a subtask and no longer using the preset classification model for division.

[0016] Furthermore, the information sharing subsystem also includes a device information sharing module; the device information sharing module is used to collect personal device parameters equipped by each employee in the enterprise, as well as common device parameters shared by each employee in the enterprise.

[0017] Furthermore, the task analysis unit is also used to quantify the equipment carrying capacity required for each subtask and attach a first equipment condition label to each subtask; the personnel analysis unit is also used to quantify the equipment carrying capacity of each employee based on personal equipment parameters and attach a second equipment condition label to each employee. At the same time, the general equipment parameters are also quantified in real time to obtain a third equipment condition label; when matching subtasks with employees, the matching unit also uses the first equipment condition label as a benchmark to match the second equipment condition label and the third equipment condition label.

[0018] Option 2

[0019] A general multifunctional decision-making method, using a general multifunctional decision-making system as described in Solution 1 to assist in enterprise management decision-making, comprises the following steps:

[0020] The personal information sharing module collects employees' personal task information and allows employees to upload their personal task information; the inter-group information sharing module collects personnel arrangement information, task arrangement information, and historical project information of each department and group in the enterprise;

[0021] The task collection module collects tasks to be assigned; the task analysis unit divides the tasks to be assigned by time sequence and by section, and obtains multiple subtasks in time sequence; the personnel analysis unit determines the employee's task acceptance time zone and task acceptance type based on personal task information and data collected by the inter-group information sharing module; the matching unit matches subtasks with employees in time sequence based on the analysis results of the task analysis unit and the personnel analysis unit, and generates task allocation suggestions.

[0022] The working principle and advantages of the present invention are:

[0023] The present invention provides a universal multifunctional decision-making system and method, which can quantitatively analyze the task carrying capacity of employees in combination with their actual conditions, and can intuitively display the task processing capabilities and adapted task types of each employee. In addition, this solution can also automatically collect tasks to be assigned and parse the tasks, and match the parsed subtasks with employees. It can use intelligent data analysis to set task allocation plans for enterprises, which helps to achieve synergy and efficiency in enterprise work. In particular, firstly, before the matching decision process, this solution specially sets up a preset neural network model to mine and quantify the task carrying capacity, which can provide objective and reliable data reference for matching decisions; in the matching decision process, this solution sets up a multi-level matching strategy, including from the initial matching pair to the target matching pair, and advanced screening of the target matching pair, which can analyze and obtain the employee who is most suitable for the subtask, and ensure that the task carrying capacity of each employee is not overloaded, and the task allocation decision is more reasonable and effective.

[0024] In addition, when making task allocation decisions, this solution also collects and analyzes device information and quantifies it into device carrying capacity. This can further consider the impact of device configuration conditions on task processing capabilities and further optimize task allocation decisions. Among them, this solution breaks the ground by considering that in the context of big data, the execution of work in many enterprises (especially those related to data analysis and processing) is often closely related to cloud resources and hardware resources. (For example, the execution of a data analysis task requires not only employee participation, but also cloud servers and computer hardware to provide computing power support. Cloud and hardware resources have become a key link in task execution, and sufficient cloud and hardware resources are required to support the efficient execution of tasks.) Based on this, this solution not only quantifies the task carrying capacity of individual employees, but also quantifies the equipment carrying capacity of their equipped conditions. This combines employees and equipment into an integrated task processing unit. When making task allocation decisions, task allocation recommendations are made based on the quantitative assessment of the task carrying capacity of the entire task processing unit. This makes the recommendation formulation more objective and comprehensive, and can provide more practical and objective decision-making recommendations for enterprises in the context of big data. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a structural diagram of a general multifunctional decision-making system and method according to a first embodiment of the present invention;

[0026] Figure 2 This is a structural diagram of a second embodiment of a general multifunctional decision-making system and method of the present invention. DETAILED DESCRIPTION

[0027] The following is a further detailed description through specific implementation methods.

[0028] Example 1

[0029] The embodiment is basically as shown in the attached Figure 1 Shown: A general multifunctional decision-making system, including an information sharing subsystem and a decision-making subsystem.

[0030] The information sharing subsystem includes a personal information sharing module and an inter-group information sharing module.

[0031] The personal information sharing module is used to automatically collect employees' personal task information and allow employees to upload personal task information. Specifically, the personal task information includes the employee's current task data and completed task data; the current task data includes the planned completion time of the task, basic task information (such as task name, description, objectives, expected results, etc.), task-related attachments, etc. The completed task data includes basic task information, time management data (including task start date, deadline, progress tracking, and extension status), responsibility allocation data (including task leader, team members, their respective responsibilities and roles, etc.), progress status data (including the current completion status of the task, completion status of each stage, progress percentage, etc.) and the working hours spent by the employee on the completed task.

[0032] The inter-group information sharing module collects personnel and task scheduling information, as well as historical project information, from each department within the enterprise. This information, combined with inter-group information, can be used by subsequent personnel analysis units to further identify the tasks each employee is actually responsible for, helping to improve the accuracy of task capacity analysis.

[0033] The decision-making subsystem includes a task collection module and a task decision module.

[0034] The task collection module is used to collect pending tasks. In this embodiment, the task collection module can establish a two-way information connection with the enterprise's PM system (project management system) and OA system, and capture pending task information from the PM system and OA system. Employees can also upload pending tasks to the task collection module.

[0035] The task decision module includes a task analysis unit, a personnel analysis unit and a matching unit.

[0036] The task analysis unit is used to divide the tasks to be assigned by time sequence and by section, and obtain multiple subtasks in the time sequence. Specifically, the task analysis unit uses a preset classification model to divide the tasks to be assigned; the preset classification model is a CNN-RNN joint model that can complete the task division based on the semantic description of the task.

[0037] Moreover, before dividing the tasks to be assigned, an initial judgment is also made on the tasks to be assigned; the initial judgment includes: searching for a target word in the data of the tasks to be assigned, and if the search result shows that the target word matches successfully and the number of target word matches is less than 2, then the task to be assigned is directly treated as a subtask, and the preset classification model is no longer used for division. In actual applications, the target word can be customized based on the task situation of the enterprise. In this way, for some daily small tasks (such as filling out a weekly report, etc.), the task division can be completed directly based on the preset target word, without the need to split it into redundant subtasks, which can reduce the data processing volume of the task analysis unit and avoid excessive task splitting and prolonging the workflow.

[0038] The personnel analysis unit is used to determine the employee's task acceptance time zone and task acceptance type based on personal task information and the data collected by the inter-group information sharing module. The task acceptance time zone refers to the period during which the employee can continue to accept tasks, and the task acceptance type is extracted from the completed task data. In this embodiment, the task acceptance type can be divided into information collection type, project planning type, architecture design type, project development type, project testing type, project procurement type, resource management type, risk management type, quality management type, process management type, cost management type, etc. according to the project section. Each task acceptance type can correspond to a series of subtasks.

[0039] The personnel analysis unit is further configured to quantify the task carrying capacity of each employee based on the employee's completed task data.

[0040] The matching unit is used to match subtasks with employees in chronological order according to the analysis results of the task analysis unit and the personnel analysis unit, and generate task allocation suggestions.

[0041] When matching subtasks with employees in chronological order, the following steps are included: based on the chronological order of the subtasks, first match them with the employee's task acceptance time zone, and filter to obtain an initial matching pair; then, based on the initial matching pair, match them with the employee's task acceptance type, and filter to obtain a target matching pair.

[0042] The initial matching pairs are many-to-many (i.e., multiple subtasks are matched with multiple employees), while the target matching pairs are one-to-many (i.e., one subtask is matched with multiple employees). After selecting the target matching pairs, the task capacity occupied by the subtasks is evaluated, and the employee with the largest task capacity margin is selected as the preferred employee to match the subtask; the remaining employees are selected as backup candidates. The generated task assignment suggestions include each subtask of the task to be assigned and its corresponding preferred and backup employees.

[0043] When quantifying the task carrying capacity of each employee and evaluating the task carrying capacity occupied by the subtasks, a preset neural network model is used for quantitative evaluation.

[0044] The preset neural network model includes an input layer, a self-attention network layer, an RNN network layer, a first fully connected layer, a Relu function layer, a second fully connected layer, and an output layer. The self-attention network layer helps the network accurately capture data features. The RNN network layer can quantify task capacity by combining temporal relationships. The Relu layer is used to introduce nonlinear relationships, which increases the network's expressive power and enables accurate analysis of complex data.

[0045] When quantifying each employee's task carrying capacity, the preset neural network model takes as input the employee's completed task data and outputs the threshold value of the employee's task carrying capacity. When evaluating the task carrying capacity occupied by a subtask, the preset neural network model takes as input the employee's completed task data and subtask data, and outputs the proportion of the subtask in the employee's task carrying capacity. Furthermore, when performing quantitative assessments, the preset neural network's self-attention network focuses on capturing the employee's work hours spent on the completed task.

[0046] In this embodiment, the preset neural network model is also trained using a historical data set, which contains multiple different types of completed tasks (such as business type tasks, daily affairs management tasks, etc.) and the average working hours occupied by the task (this working hours can be obtained using statistical methods).

[0047] This embodiment also provides a general multifunctional decision-making method, which uses a general multifunctional decision-making system as described above to assist in enterprise management decision-making; the method includes the following steps:

[0048] The personal information sharing module collects employees' personal task information and allows employees to upload their personal task information; the inter-group information sharing module collects personnel arrangement information, task arrangement information, and historical project information of each department and group in the enterprise;

[0049] The task collection module collects tasks to be assigned; the task analysis unit divides the tasks to be assigned by time sequence and by section, and obtains multiple subtasks in time sequence; the personnel analysis unit determines the employee's task acceptance time zone and task acceptance type based on personal task information and data collected by the inter-group information sharing module; the matching unit matches subtasks with employees in time sequence based on the analysis results of the task analysis unit and the personnel analysis unit, and generates task allocation suggestions.

[0050] This embodiment provides a universal multifunctional decision-making system and method that can intelligently set and optimize enterprise task allocation plans based on the actual situation of employees. It is rich in functions and highly versatile, and helps to achieve synergy and efficiency in enterprise work.

[0051] Example 2

[0052] As attached Figure 2 As shown, a general multifunctional decision-making system, based on embodiment 1, the decision-making subsystem also includes an adjustment module; the adjustment module is used to monitor the task execution status of the employees corresponding to each subtask in real time. If the corresponding employee is unable to perform the subtask (such as employee leave, employee transfer, etc.), the task decision module is immediately linked to regenerate the task allocation suggestion for the subtask.

[0053] Compared with the first embodiment, the general multifunctional decision-making system and method provided in this embodiment can timely update the task allocation suggestions based on the actual execution status of the task, and can realize follow-up decision-making.

[0054] Example 3

[0055] A general multifunctional decision-making system, based on the first embodiment, further fine-tunes the percentage of a subtask's task carrying capacity, based on the employee's proficiency in the subtask, when evaluating the subtask's task carrying capacity. Specifically, after obtaining the percentage of the subtask in the employee's task carrying capacity as output by a preset neural network model, the percentage is fine-tuned based on the employee's proficiency in the subtask. The fine-tuning comprises: multiplying the percentage by the proficiency, and obtaining a new percentage.

[0056] If there is no task data consistent with the subtask in the employee's historical task data, the task proficiency is determined to be 1, and no fine-tuning is done. If there is task data consistent with the subtask in the employee's historical task data, the task proficiency is evaluated according to the historical completion times and completion quality of this type of subtask. In this embodiment, task proficiency = 1-(Cn+Dm)%; C and D are weight coefficients, n is the historical completion times, and m is the completion quality; the more historical completion times and the higher the completion quality, the smaller the value of the task proficiency, and the smaller the proportion of the corresponding subtask in the employee's task carrying capacity.

[0057] The general multifunctional decision-making system and method provided in this embodiment, compared with the first embodiment, has a more detailed assessment of task carrying capacity, can more accurately measure the adaptation relationship between different subtasks and different employees, and make more reasonable decisions.

[0058] Example 4

[0059] A general multifunctional decision-making system, based on embodiment 1, the information sharing subsystem also includes a device information sharing module; the device information sharing module is used to collect personal device parameters equipped by each employee in the enterprise, as well as common device parameters shared by all employees in the enterprise.

[0060] In this embodiment, the personal device parameters include the CPU performance, GPU performance, memory size, hard disk read and write speed, storage capacity, network bandwidth, delay time, key software version (the key software version here can be specified according to the actual needs of the enterprise, such as Python 3.10, SQL Server 2022), number of licenses, interface type, technical support response time, etc. of the office equipment (such as a computer) equipped by the employee.

[0061] Through personal equipment parameters, we can intuitively analyze the equipment conditions of each employee, which helps to more comprehensively and objectively analyze their comprehensive carrying capacity when handling tasks.

[0062] The general equipment parameters include cloud database throughput, cloud database response time, cloud database storage capacity, cloud database availability (here refers to the uptime percentage of cloud database services, usually given in the form of SLA), RTO (here refers to an indicator for measuring disaster recovery speed, which indicates the time required from failure to recovery to normal operation; the shorter the RTO, the faster it can recover to normal operation in the event of a failure), RPO (here refers to an indicator for measuring disaster recovery capability, which indicates the time point to which the most recent available backup can be recovered in the event of a catastrophic failure; the higher the RPO, the closer the time point to which it can be recovered in the event of a failure), network packet loss rate, server CPU utilization, server memory utilization, maximum number of concurrent connections, number of concurrent users of collaboration tools, file synchronization speed, firewall effectiveness, etc.

[0063] Through common device parameters, the performance and stability of shared devices in the enterprise can be intuitively analyzed, thereby better quantifying the processing conditions for employees' data processing and collaboration needs.

[0064] The task analysis unit is further configured to quantify the equipment capacity required for each subtask and assign a first equipment condition tag to each subtask. The personnel analysis unit is further configured to quantify the equipment capacity of each employee based on their individual equipment parameters and assign a second equipment condition tag to each employee. Furthermore, the matching unit performs real-time quantification of general equipment parameters (i.e., quantifies the equipment capacity of general equipment) and assigns a third equipment condition tag. When matching subtasks with employees, the matching unit uses the first equipment condition tag as a benchmark to match the second and third equipment condition tags.

[0065] In this embodiment, the task analysis unit and the personnel analysis unit each use an MLP (multi-layer perceptron) network based on the tensorflow library to perform quantitative analysis when quantifying the equipment carrying capacity.

[0066] To quantify the carrying capacity of each employee's device, the MLP network used to quantify individual device parameters first collects historical individual device parameters as training and validation sets for network training and validation. During training, historical individual device parameter data is used as input, and the target variable (i.e., output) is set to a comprehensive score, i.e., the carrying capacity of the device. Based on the trained MLP network, the carrying capacity of each employee's device is then quantified.

[0067] When quantifying the equipment carrying capacity required for each subtask, the corresponding MLP network uses an improved historical dataset, divided into a training set and a validation set, for network training and validation. The improved historical dataset contains the data from the historical dataset in Example 1, as well as historical personal equipment parameter data for completed tasks. During training, the various data points in the improved historical dataset are used as input, and the target variable (i.e., output) is set to a comprehensive score, namely, the equipment carrying capacity. Based on the trained MLP network, the equipment carrying capacity required for each subtask is quantified.

[0068] When quantifying general equipment parameters, the corresponding MLP network first collects historical general equipment parameters as training and validation sets for network training and validation. During training, the historical general equipment parameter data is used as input, and the target variable (i.e., output) is set to a comprehensive score, namely the equipment carrying capacity. Based on the trained MLP network, the general equipment parameters are then quantified.

[0069] When matching subtasks with employees, the matching unit first uses the first device condition tag as a benchmark and matches it with the second device condition tag. If the task carrying capacity in the second device condition tag is greater than the task carrying capacity in the first device condition tag, then the subtask corresponding to the first device condition tag is determined to be matched with the employee corresponding to the second device condition tag to form a matching pair, otherwise they are not matched. If they are matched, they are matched with the third device condition tag. If the task carrying capacity in the third device condition is greater than the task carrying capacity in the first device condition tag, and the number of concurrent connections of general devices and the number of concurrent users of collaboration tools under the conditions for this matching pair to be established are both less than the threshold, then this matching pair is established, otherwise it is not established.

[0070] In specific applications, the matching unit may introduce the first device condition tag, the second device condition tag, and the third device condition tag on the basis of the initial matching pair or the target matching pair to perform advanced matching; or first perform matching based on the first device condition tag, the second device condition tag, and the third device condition tag, and then screen the initial matching pair and the target matching pair, and output task allocation suggestions based on the final matching pair; this setting can reduce the overall matching analysis amount while ensuring the accuracy of the matching.

[0071] Compared with the first embodiment, the general multifunctional decision-making system and method provided in this embodiment can further consider the impact of equipment configuration conditions on task processing capabilities when making task allocation decisions; it is particularly suitable for enterprises where work efficiency is highly correlated with hardware configuration conditions, such as big data analysis enterprises, information service enterprises, etc.

[0072] The above is only an embodiment of the present invention. Common knowledge such as the specific structure and characteristics of the scheme is not described in detail here. Ordinary technicians in the relevant field are aware of all common technical knowledge in the technical field of the invention before the application date or priority date, can obtain all existing technologies in the field, and have the ability to apply conventional experimental means before that date. Ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the guidance of this application. Some typical well-known structures or well-known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several variations and improvements can be made, which should also be regarded as the scope of protection of the present invention. These will not affect the effect of the implementation of the present invention and the practicality of the patent.

Claims

1. A general multifunctional decision-making system, characterized in that: It includes an information sharing subsystem and a decision-making subsystem; the information sharing subsystem includes a personal information sharing module and an inter-group information sharing module; the personal information sharing module is used to automatically collect employees' personal task information and allow employees to upload their personal task information; the inter-group information sharing module is used to collect personnel arrangement information, task arrangement information and historical project information of each department and group in the enterprise; The decision-making subsystem includes a task collection module and a task decision module; the task collection module is used to collect tasks to be assigned; the task decision module includes a task analysis unit, a personnel analysis unit and a matching unit; the task analysis unit is used to divide the tasks to be assigned according to time sequence and by section, and obtain multiple subtasks in time sequence; the personnel analysis unit is used to determine the employee's task acceptance time zone and task acceptance type based on personal task information and data collected by the inter-group information sharing module; the matching unit is used to match subtasks with employees in time sequence based on the analysis results of the task analysis unit and the personnel analysis unit, and generate task allocation suggestions; The personal task information includes the employee's current task data and completed task data; the current task data includes the planned completion time of the task; the personnel analysis unit is further used to quantify the task carrying capacity of each employee based on the employee's completed task data; When matching subtasks with employees by time sequence, the following steps are included: based on the time sequence of the subtasks, first matching with the employee's task acceptance time zone, and screening to obtain an initial matching pair; then, based on the initial matching pair, matching with the employee's task acceptance type, and screening to obtain a target matching pair; The initial matching pair is a many-to-many match, and the target matching pair is a one-to-many match. After the target matching pair is obtained through screening, the task carrying capacity occupied by the subtask is evaluated, and the employee with the larger task carrying capacity margin is selected to match the subtask. When quantifying the task carrying capacity of each employee and evaluating the task carrying capacity occupied by subtasks, a preset neural network model is used for quantitative evaluation; After obtaining the proportion of the subtask in the employee's task carrying capacity output by the preset neural network model, the proportion is further fine-tuned based on the employee's task proficiency in the subtask; the fine-tuning includes: , and get the new proportion; If there is no task data matching the subtask in the employee's historical task data, the task proficiency is determined to be 1, and no fine-tuning is performed. If there is task data matching the subtask in the employee's historical task data, the task proficiency is evaluated based on the historical number of completions and completion quality of the subtask. C and D are weight coefficients, n is the number of historical completions, and m is the completion quality. The more historical completions and the higher the completion quality, the smaller the value of task proficiency, and the smaller the proportion of the corresponding subtask in the employee's task carrying capacity. The information sharing subsystem also includes a device information sharing module; the device information sharing module is used to collect personal device parameters equipped by each employee in the enterprise, as well as common device parameters shared by all employees in the enterprise; the task analysis unit is also used to quantify the device carrying capacity required for each subtask and attach a first device condition tag to each subtask; the personnel analysis unit is also used to quantify the device carrying capacity equipped by each employee based on the personal device parameters, and attach a second device condition tag to each employee, and at the same time, quantify the common device parameters in real time and obtain a third device condition tag; when matching subtasks with employees, the matching unit also uses the first device condition tag as a reference to match the second device condition tag and the third device condition tag; When quantifying the equipment carrying capacity, the task analysis unit and the personnel analysis unit respectively use an MLP network based on the tensorflow library for quantitative analysis; when matching subtasks with employees, the matching unit first uses the first equipment condition label as a benchmark and matches it with the second equipment condition label. If the task carrying capacity in the second equipment condition label is greater than the task carrying capacity in the first equipment condition label, then it is determined that the subtask corresponding to the first equipment condition label is matched with the employee corresponding to the second equipment condition label to form a matching pair, otherwise they are not matched; if they are matched, they are matched with the third equipment condition label. If the task carrying capacity in the third equipment condition is greater than the task carrying capacity in the first equipment condition label, and the number of concurrent connections of general devices and the number of concurrent users of collaboration tools under the conditions for this matching pair to be established are both less than the threshold, then this matching pair is established, otherwise it is not established; Based on the initial matching pair or the target matching pair, the matching unit introduces the first device condition tag, the second device condition tag, and the third device condition tag to perform advanced matching; or first performs matching based on the first device condition tag, the second device condition tag, and the third device condition tag, and then screens the initial matching pair and the target matching pair, and outputs a task allocation suggestion based on the final matching pair.

2. A general multifunctional decision-making system according to claim 1, characterized in that: The preset neural network model is provided with an input layer, a self-attention network layer, an RNN network layer, a first fully connected layer, a Relu function layer, a second fully connected layer and an output layer.

3. A general multifunctional decision-making system according to claim 2, characterized in that: When quantifying the task carrying capacity of each employee, the preset neural network model takes as input the completed task data of the employee and outputs the threshold value of the employee's task carrying capacity; when evaluating the task carrying capacity occupied by a subtask, the preset neural network model takes as input the completed task data and subtask data of the employee and outputs the proportion of the subtask in the employee's task carrying capacity.

4. A general multifunctional decision-making system according to claim 1, characterized in that: When dividing the tasks to be assigned, the task analysis unit adopts a preset classification model for division; the preset classification model is a CNN-RNN joint model; and before dividing the tasks to be assigned, an initial judgment is also made on the tasks to be assigned; the initial judgment includes: searching for target words in the data of the tasks to be assigned, and if the search result shows that the number of target word matches is less than 2, then directly treating the task to be assigned as a subtask and no longer using the preset classification model for division.

5. A general multifunctional decision-making method, characterized in that: A general multifunctional decision-making system according to any one of claims 1 to 4 is used to assist in enterprise management decision-making; the system comprises the following steps: The personal information sharing module collects employees' personal task information and allows employees to upload their personal task information; the inter-group information sharing module collects personnel arrangement information, task arrangement information, and historical project information of each department and group in the enterprise; The task collection module collects tasks to be assigned; the task analysis unit divides the tasks to be assigned by time sequence and by section, and obtains multiple subtasks in time sequence; the personnel analysis unit determines the employee's task acceptance time zone and task acceptance type based on personal task information and data collected by the inter-group information sharing module; the matching unit matches subtasks with employees in time sequence based on the analysis results of the task analysis unit and the personnel analysis unit, and generates task allocation suggestions.

Images