Resource putting method and device based on dynamic regulation, storage medium and computer device

Abstract

This application can be applied to fields such as resource allocation, fintech, and smart healthcare. It discloses a resource allocation method and apparatus, storage medium, and computer equipment based on dynamic control. The method includes: acquiring the resource budget plan and project execution data of the target project; calculating and predicting resource demand based on the project execution data using a preset project resource prediction model; if the resource budget plan covers the predicted resource demand, calculating a resource idle matrix and matrix score based on the resource budget plan and the predicted resource demand; determining a first budget adjustment plan for the resource budget plan based on the matrix score; if the resource budget plan does not cover the predicted resource demand, acquiring multiple project nodes of the target project and the expected targets of each project node; calculating the predicted workload value based on the expected node targets using a preset workload prediction model; and determining a second budget adjustment plan for the resource budget plan based on the predicted workload value of each project node.

Description

Technical Field

[0001] This application relates to the field of data processing technology and can be applied to fields such as resource allocation, fintech, and smart healthcare. In particular, it relates to a resource allocation method and device based on dynamic control, a storage medium, and a computer device. Background Technology

[0002] In today's era of rapid technological advancement, digital transformation has become an essential path for various industries to enhance competitiveness and achieve sustainable development. As a crucial component of economic development, the financial sector, including the leasing industry, faces an increasingly urgent need for digital transformation in its technology resource management. Currently, the financial sector is leveraging technology to continuously expand its service boundaries and enhance risk control capabilities, while the smart healthcare technology field is optimizing treatment processes and improving medical quality through digital means. However, both the efficient allocation of technology resources in the financial sector and the rational utilization of medical equipment and information resources in the smart healthcare field face the challenge of urgently needing to innovate their technology resource management methods. Traditional management models are no longer adequate to adapt to the rapidly changing market environment and business needs; therefore, the intelligent and systematic transformation of technology resource management is imperative.

[0003] Currently, technology resource management systems in various fields, such as finance and smart healthcare, primarily rely on traditional decentralized management models, lacking dynamic and refined management systems. Specifically, in budget preparation, resource budget plans for projects are typically determined manually based on experience. This approach is not only subjective and uncertain, but also lacks an effective dynamic adjustment mechanism during the execution of the budget plan, which strictly adheres to the predetermined schedule. The lack of technical means to effectively analyze and predict project execution data prevents intelligent dynamic control, frequently resulting in inaccurate resource budgets or unreasonable resource allocation during actual implementation. Summary of the Invention

[0004] In view of this, this application provides a resource allocation method and apparatus, storage medium, and computer equipment based on dynamic control. It improves the accuracy of resource demand forecasting through the analysis of project execution data; by constructing a resource idleness matrix and calculating matrix scores, it transforms resource idleness into quantifiable and comparable technical indicators, providing precise data-driven basis for resource adjustment; and through node workload prediction and resource coverage analysis, it realizes intelligent and refined resource allocation planning under resource constraints. This application enables dynamic and refined management and precise allocation of scientific and technological resources, effectively avoiding problems of insufficient budget and resource idleness, significantly improving the data processing capabilities and intelligent decision-making level of the resource management system, and contributing to the optimization of resource allocation.

[0005] According to one aspect of this application, a resource allocation method based on dynamic control is provided, comprising: Obtain the resource budget plan and project execution data corresponding to the target project, and calculate the predicted resource requirements of the target project based on the project execution data using a preset project resource prediction model; If the resource budget plan covers the predicted resource demand, then a resource idle matrix is ​​calculated based on the resource budget plan and the predicted resource demand, and a matrix score corresponding to the resource idle matrix is ​​calculated. A first budget adjustment plan corresponding to the resource budget plan is dynamically determined based on the matrix score, so as to allocate resources based on the first budget adjustment plan. If the resource budget plan does not cover the predicted resource demand, then obtain multiple project nodes corresponding to the target project, and the node expected target corresponding to each project node. Based on the node expected target corresponding to each project node, calculate the predicted workload value corresponding to each project node through a preset workload prediction model. Based on the predicted workload value corresponding to each project node, dynamically determine the second budget adjustment plan corresponding to the resource budget plan, and allocate resources based on the second budget adjustment plan.

[0006] According to another aspect of this application, a resource allocation device based on dynamic control is provided, comprising: The prediction module is used to obtain the resource budget plan and project execution data corresponding to the target project, and calculate the predicted resource requirements of the target project based on the project execution data through a preset project resource prediction model. The first adjustment module is used to calculate a resource idle matrix based on the resource budget plan and the predicted resource demand if the resource budget plan covers the predicted resource demand, and to calculate the matrix score corresponding to the resource idle matrix. Based on the matrix score, the module dynamically determines a first budget adjustment plan corresponding to the resource budget plan, so as to allocate resources based on the first budget adjustment plan. The second adjustment module is used to obtain multiple project nodes corresponding to the target project and the expected target of each project node if the resource budget plan does not cover the predicted resource demand. Based on the expected target of each project node, the module calculates the predicted workload value of each project node through a preset workload prediction model. Based on the predicted workload value of each project node, the module dynamically determines the second budget adjustment plan corresponding to the resource budget plan, and allocates resources based on the second budget adjustment plan.

[0007] According to another aspect of this application, a storage medium is provided that stores a computer program thereon, which, when executed by a processor, implements the above-described resource allocation method based on dynamic control.

[0008] According to another aspect of this application, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor executes the program to implement the above-described resource allocation method based on dynamic control.

[0009] By employing the above technical solutions, this application provides a resource allocation method and apparatus, storage medium, and computer equipment based on dynamic control. First, it acquires the resource budget plan and project execution data corresponding to the target project. Then, it calculates and predicts resource demand using a preset project resource prediction model. Based on the relationship between the resource budget plan and the predicted resource demand, different strategies are adopted to determine the budget adjustment plan: when the resource budget plan covers the predicted resource demand, a first budget adjustment plan is determined by calculating the resource idle matrix and its score; when the resource budget plan does not cover the predicted resource demand, a second budget adjustment plan is determined based on the node expected targets and workload prediction values ​​of the project nodes. This method improves the accuracy of resource demand prediction through the analysis of project execution data; by constructing a resource idle matrix and calculating matrix scores, the resource idle status is transformed into quantifiable and comparable technical indicators, providing precise data-driven basis for resource adjustment; through node workload prediction and resource coverage analysis, intelligent and refined resource allocation planning under resource constraints is realized. The embodiments of this application can achieve dynamic and refined management and precise allocation of scientific and technological resources, effectively avoiding the problems of insufficient budget and resource idleness, significantly improving the data processing capabilities and intelligent decision-making level of the resource management system, and contributing to the optimization of resource allocation.

[0010] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0011] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 A flowchart illustrating a resource allocation method based on dynamic control provided in an embodiment of this application is shown. Figure 2 A schematic diagram of a resource allocation device based on dynamic control provided in an embodiment of this application is shown. Figure 3 A schematic diagram of the device structure of a computer device provided in an embodiment of this application is shown. Detailed Implementation

[0012] The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present application can be combined with each other.

[0013] This embodiment provides a resource allocation method based on dynamic control, such as... Figure 1 As shown, the method includes: Step 101: Obtain the resource budget plan and project execution data corresponding to the target project, and calculate the predicted resource requirements of the target project based on the project execution data using a preset project resource prediction model.

[0014] Step 102: If the resource budget plan covers the predicted resource demand, then calculate the resource idle matrix based on the resource budget plan and the predicted resource demand, and calculate the matrix score corresponding to the resource idle matrix. Dynamically determine the first budget adjustment plan corresponding to the resource budget plan based on the matrix score, so as to allocate resources based on the first budget adjustment plan.

[0015] Step 103: If the resource budget plan does not cover the predicted resource demand, then obtain multiple project nodes corresponding to the target project, and the node expected target corresponding to each project node. Based on the node expected target corresponding to each project node, calculate the predicted workload value corresponding to each project node through a preset workload prediction model. Based on the predicted workload value corresponding to each project node, dynamically determine the second budget adjustment plan corresponding to the resource budget plan, and allocate resources based on the second budget adjustment plan.

[0016] This application provides a resource allocation method based on dynamic control, which can meet the needs of the financial sector, such as the leasing industry, and the smart healthcare technology sector for precise allocation, dynamic adjustment, and efficient utilization of technological resources. First, a predetermined resource budget plan for the target project is obtained. This plan clarifies key information such as the total amount and allocation method of various technological resources set during the planning phase. For example, in the fintech field, the resource budget plan for a new financial product R&D project can detail the allocation of human resources (including personnel resources for different professional positions) and technical equipment resources (such as servers, data analysis software, etc.) throughout the project cycle. Simultaneously, project execution data can be collected. This data reflects the actual status of the target project during its progress. For example, in a medical information system upgrade project in the smart healthcare field, the project execution data can cover system development progress, invested man-hours, and used server resources.

[0017] Subsequently, based on the collected project execution data, a pre-defined project resource prediction model is used to calculate the project's predicted resource requirements. This prediction model can be one trained and validated with extensive historical data. It can reasonably estimate the future resource needs of the target project based on its current execution status, industry trends, and project characteristics, thus obtaining the predicted resource requirements. Taking a fintech project as an example, if the current project is in the system testing phase, based on historical test data from similar projects and the project's current testing progress, the number of defects discovered, and other execution data, the pre-defined project resource prediction model can estimate the additional manpower and time resources required for subsequent testing and defect fixing. This step provides accurate target guidance for subsequent resource adjustments.

[0018] Furthermore, it's crucial to determine whether the resource budget plan can meet the projected resource needs. If the plan does, it indicates that the target project may have some resource redundancy. In this case, a resource availability matrix can be calculated based on the budget plan and projected resource needs. The resource availability matrix can be a multi-dimensional data structure that displays resource availability across different budget dimensions (such as human, material, and financial resources). For example, in a fintech project, assuming the budget dimensions include developers, testing equipment, and funding, by comparing the budget plan and projected resource needs, the amount of resource availability for each budget dimension can be determined, thus constructing the resource availability matrix. After calculating the resource availability matrix, its corresponding matrix score can be calculated. The matrix score is a comprehensive quantitative assessment of resource availability. Then, the first budget adjustment plan corresponding to the resource budget plan can be dynamically determined based on the matrix score. Specifically, if the matrix score is high, it indicates a more severe resource availability situation. In this case, a significant adjustment to the budget plan can be made, such as reallocating idle funds to other projects urgently needing funding, or allocating idle human resources to other parallel projects. Conversely, if the matrix score is low, it indicates that resource availability is within a reasonable range, requiring only minor adjustments, such as optimizing resource usage schedules to ensure fuller utilization of resources throughout the project cycle. Subsequently, resource allocation can be based on the determined initial budget adjustment plan to achieve rational allocation and efficient utilization of technological resources.

[0019] When the resource budget plan cannot meet the predicted resource needs, it indicates a resource gap in the target project. In this case, the first step is to obtain multiple project nodes corresponding to the target project and the expected goals for each node. Project nodes are key milestones in the project's progress. For example, in a fintech project launching a new financial product, project nodes may include completion of requirements gathering, approval of system design review, completion of development, and successful testing. Each project node has clear expected goals; for example, the completion of requirements gathering requires clarifying key information such as the product's functional requirements and target user group. Then, based on the expected goals of each project node, a pre-defined workload prediction model is used to calculate the predicted workload for each node. The workload prediction model can consider factors such as the complexity of the project node, the required skill level, and workload data from similar historical projects. Taking a telemedicine system construction project in the smart healthcare field as an example, for the project node of system development completion, the workload prediction model can predict the workload indicators such as man-hours and lines of code required to complete this project node based on the number of functional modules in the system, the technical difficulty (such as whether it involves artificial intelligence algorithms or big data analysis), the technical capabilities of the development team, and the workload data from similar historical telemedicine systems.

[0020] Next, based on the predicted workload for each project node, a second budget adjustment plan can be dynamically determined corresponding to the existing resource budget plan. For example, it can be determined which project nodes the existing resource budget plan can cover, and these project nodes can be specified in the second budget adjustment plan. Then, resources can be allocated based on the determined second budget adjustment plan to complete the preceding project nodes with limited resources as much as possible, ensuring the smooth implementation of the target project.

[0021] By applying the technical solution of this embodiment, the resource budget plan and project execution data corresponding to the target project are first obtained. Then, a preset project resource prediction model is used to calculate the predicted resource demand. Based on the relationship between the resource budget plan and the predicted resource demand, different strategies are adopted to determine the budget adjustment plan: when the resource budget plan covers the predicted resource demand, the first budget adjustment plan is determined by calculating the resource idle matrix and its score; when the resource budget plan does not cover the predicted resource demand, the second budget adjustment plan is determined based on the node expected targets and workload prediction values ​​of the project nodes. This method improves the accuracy of resource demand prediction by analyzing project execution data; by constructing a resource idle matrix and calculating the matrix score, the resource idle status is transformed into quantifiable and comparable technical indicators, providing a precise data-driven basis for resource adjustment; through node workload prediction and resource coverage analysis, intelligent and refined resource allocation planning under resource constraints is realized. This embodiment of the application can realize dynamic and refined management and precise allocation of scientific and technological resources, effectively avoid the problems of insufficient budget and idle resources, significantly improve the data processing capability and intelligent decision-making level of the resource management system, and help optimize resource allocation.

[0022] Optionally, in this embodiment, the preset project resource prediction model performs the following operations to obtain the predicted resource requirements of the target project: extracting structured and unstructured data from the project execution data; determining project characteristic indicator data corresponding to the target project based on the structured data, and inputting the project characteristic indicator data into a time series analysis sub-model and a regression analysis sub-model respectively, obtaining a first predicted requirement through the time series analysis sub-model, and obtaining a second predicted requirement through the regression analysis sub-model; performing data mining on the unstructured data to obtain supplementary project indicator data, and inputting the supplementary project indicator data into a demand prediction agent, obtaining a third predicted requirement through the demand prediction agent; and determining the predicted resource requirements of the target project based on the first predicted requirement, the second predicted requirement, and the third predicted requirement.

[0023] In this embodiment, the workflow of the preset project resource prediction model is as follows: First, extract structured and unstructured data from project execution data. Project execution data can include a wealth of content generated during the project's progress. Structured data refers to data with a fixed format and structure. For example, in a loan approval system development project in the fintech field, structured data could include the number of developers involved, the working hours of each developer, the amount of funds spent on the project, and the completion percentage of each functional module of the system. This data usually exists in the form of tables or database records, making it easy to read and analyze directly. Unstructured data, on the other hand, lacks a fixed format and structure and is more diverse in form. For example, in a medical image analysis system development project in the smart healthcare field, unstructured data could include text of project meeting minutes, emails between development team members, and project-related documents (such as requirements documents and design documents). This data contains rich semantic information, but requires specific processing to extract valuable content. By extracting structured and unstructured data separately from project execution data, a foundational data support can be provided for subsequent analyses from different dimensions.

[0024] After acquiring structured data, the next step is to extract project characteristic indicators. These indicators are quantitative descriptions of the project's key features, reflecting its essential attributes. Examples include the number of developers already involved, the working hours of each developer, the amount of funding already spent, and the percentage of completion for each functional module of the system. Next, the project characteristic indicators can be input into the time series analysis sub-model and the regression analysis sub-model. The time series analysis sub-model focuses on the patterns of data changes over time, using time series analysis of historical data to predict the target project's primary demand. For example, performing time series analysis on the target project's historical human resource data can yield future human resource forecasts; or performing time series analysis on human resource data from similar projects can also yield future human resource forecasts. The regression analysis sub-model focuses on studying the relationships between different variables, using mathematical models to predict the target variable. In a fintech project, assuming the goal is to predict human resource costs during system development, the regression analysis sub-model can consider multiple variables such as the number of developers, development difficulty coefficient, and project progress, analyzing the correlation between these variables and human resource costs to predict future human resource cost requirements. The first predicted demand obtained from the time series analysis sub-model and the second predicted demand obtained from the regression analysis sub-model can predict resource demand from different perspectives, providing a reference for subsequent comprehensive judgment.

[0025] For the extracted unstructured data, data mining can be performed to extract supplementary project indicator data. Data mining is a technique for discovering potential patterns, correlations, and trends from large amounts of data. For example, in fintech projects, data mining can be performed on unstructured data such as project meeting minutes to discover important information such as target project acceptance information and supplementary project needs. Subsequently, the mined supplementary project indicator data is input into a demand forecasting agent. A demand forecasting agent is an intelligent system based on artificial intelligence technology that can learn and reason from multiple factors to predict resource needs, thus obtaining a third predicted demand.

[0026] After obtaining the first, second, and third projected demands, these three forecasts can be comprehensively considered to determine the final projected resource requirements for the target project. Since different forecasting models and methods have their advantages and limitations, a single forecast result may contain some bias. Therefore, a weighted average or other comprehensive calculation method can be used to combine the first and second projected demands, and then integrate the result with the third projected demand to obtain a comprehensive projected resource requirement. This final forecast result can more comprehensively reflect the resource requirements of the target project across different dimensions, improving the accuracy and reliability of the forecast.

[0027] Optionally, in this embodiment of the application, step 102, "calculating the resource availability matrix based on the resource budget scheme and the predicted resource demand," includes: determining the number of columns corresponding to the resource availability matrix based on the resource budget dimensions included in the resource budget scheme; for each resource budget dimension, calculating the resource availability corresponding to the resource budget dimension using the resource budget scheme and the predicted resource demand, and mapping the resource availability corresponding to each resource budget dimension to the first row of the resource availability matrix; for each resource budget dimension, determining the importance level corresponding to the resource budget dimension using a preset dimension importance level mapping table, and mapping the importance level corresponding to each resource budget dimension to the second row of the resource availability matrix.

[0028] In this embodiment, the resource budgeting scheme encompasses a detailed plan for resource investment in the target project, with the resource budget dimension being a key element representing different types of resource categories. For example, in the fintech field, taking a new financial product promotion project as an example, the resource budget dimension may include human resources (such as marketing personnel and customer service personnel), material resources (such as promotional materials and office equipment), and financial resources (such as advertising expenses and event planning expenses). In the smart healthcare field, for a hospital information system upgrade project, the resource budget dimension may involve the cost of technical personnel, the cost of purchasing hardware such as servers, and software licensing fees.

[0029] Next, the number of columns in the resource availability matrix can be determined, specifically based on the resource budget dimensions explicitly listed in the resource budget plan. Each resource budget dimension corresponds to one column in the resource availability matrix, ensuring that the matrix comprehensively and clearly displays the availability of different types of resources. After determining the number of columns, the corresponding resource availability amount can be calculated for each resource budget dimension. Specifically, the resource budget plan specifies the total budget for the target project across each resource budget dimension, while the projected resource demand is the predicted resource amount based on the actual situation of the target project. By comparing these two sets of data, the resource availability amount for each resource budget dimension can be calculated. After calculating the resource availability amount for each resource budget dimension, this data can be mapped to the first row of the resource availability matrix.

[0030] Besides resource availability, the importance of resource budget dimensions is also a crucial factor influencing resource allocation decisions. Specifically, the importance of each resource budget dimension can be determined using a pre-defined dimension importance level mapping table. This mapping table can be developed based on project characteristics and industry experience. In the fintech field, for new financial product promotion projects, marketing personnel within human resources can be assigned a higher importance level because they are the key force directly driving product market expansion; while the importance level of physical resources such as office equipment can be relatively lower. By consulting the pre-defined dimension importance level mapping table, the importance level corresponding to each resource budget dimension can be determined. These importance levels are then mapped to the second row of the resource availability matrix. In this way, the resource availability matrix not only displays the amount of available resources but also reflects the importance of each resource budget dimension. Through this method, the resource availability matrix provides more comprehensive and in-depth information support for dynamic resource control.

[0031] Optionally, in this embodiment of the application, the step 102 of "calculating the matrix score corresponding to the resource idle matrix" includes: scaling the resource idle amount under each resource budget dimension in the resource idle matrix to obtain the resource idle rate corresponding to each resource idle amount; for each resource budget dimension in the resource idle matrix, multiplying the resource idle rate and importance level under the resource budget dimension to obtain the product result under the resource budget dimension; and adding the product results under each resource budget dimension to obtain the matrix score corresponding to the resource idle matrix.

[0032] In this embodiment, when calculating the score of the resource idle matrix, the resource idle amounts with different dimensions and numerical ranges under different resource budget dimensions are first standardized to make them comparable. For example, for an online payment system development project in the fintech field, multiple resource budget dimensions are involved, such as human resources (e.g., the number of developers) and financial resources (e.g., the remaining amount of the project budget). Assume that the resource idle amount is 5 people in the human resources dimension and 500,000 yuan in the financial resources dimension. The dimensions and orders of magnitude of these two values ​​differ greatly, making direct comparison and analysis difficult. Therefore, a scaling method can be used: subtract the minimum idle amount for each resource idle amount from the minimum idle amount for that dimension, and then divide by the difference between the maximum and minimum idle amounts for that dimension to obtain a value between 0 and 1. This value is the resource idle rate. Through this processing, the resource idle amounts for different resource budget dimensions are transformed to the same numerical range, facilitating subsequent comprehensive calculation and analysis. For example, in the aforementioned fintech project, after scaling, the idle capacity of 5 people under the human resources dimension can be converted into a resource idle rate of 0.2, while the idle capacity of 500,000 yuan under the financial resources dimension may be converted into a resource idle rate of 0.3. In this way, the two can be compared and calculated on the same scale.

[0033] Resource idle rate reflects the degree of resource availability across each resource budget dimension, but the importance of different resource budget dimensions varies within a project. For example, in a smart healthcare project, the stability and performance of server hardware directly impact the operational efficiency and data security of the electronic medical record system, thus having a high importance level; while some non-core software plugins have a lower importance level. The resource idle rate and importance level are then multiplied to obtain the product result for each resource budget dimension. This product result comprehensively considers both resource idleness and importance, more accurately reflecting the potential impact of each resource budget dimension on project resource allocation decisions.

[0034] Furthermore, the product results from each resource budget dimension can be summed to obtain the matrix score corresponding to the resource idleness matrix. This matrix score can comprehensively reflect the idle status of the target project across various resource budget dimensions and the importance of these resources to the target project. If the matrix score is high, it indicates that the target project has a significant amount of idle resources across multiple important resource dimensions, suggesting that resource utilization efficiency needs improvement, and these resources could be allocated to other projects or optimized. Conversely, if the matrix score is low, it indicates that the target project has limited idle resources across key resource dimensions, requiring close monitoring of resource usage to ensure the project's smooth operation.

[0035] In this embodiment of the application, optionally, step 102, "dynamically determining the first budget adjustment scheme corresponding to the resource budget scheme based on the matrix score," includes: if the matrix score is greater than a preset score threshold, obtaining the maximum idle rate corresponding to each importance level; filtering out target resource budget dimensions from the resource idle matrix whose resource idle rate is greater than the corresponding maximum idle rate; calculating the resource adjustment amount corresponding to the target resource budget dimension based on the resource idle rate and the maximum idle rate corresponding to the target resource budget dimension; obtaining resource-to-be-replenished projects from the currently executing projects, and determining the resource-to-be-replenished projects based on their urgency, resource-to-be-replenished data, and the... The resource adjustment amount corresponding to the target resource budget dimension is used to determine the target supplementary project corresponding to the resource adjustment amount; the resource budget plan is adjusted according to the resource adjustment amount corresponding to the target resource budget dimension, and a first budget adjustment plan is generated based on the adjusted resource budget plan, the resource adjustment amount corresponding to the target resource budget dimension, and the target supplementary project; if the matrix score is less than or equal to the preset score threshold, the resource idle amount corresponding to each resource budget dimension is determined according to the resource idle matrix, the resource idle amount is marked as candidate idle resources, and a first budget adjustment plan is generated based on the marked candidate idle resources and the resource budget plan.

[0036] In this embodiment, when the matrix score exceeds a preset score threshold, it indicates that the target project has a certain degree of idle resources in its overall resource utilization. Here, the preset score threshold can be a critical value set based on the actual situation of the target project and industry experience, used to determine whether the degree of resource idleness meets the standard for budget adjustment. Specifically, the maximum idle rate corresponding to each importance level can be obtained, thus clarifying the upper limit of tolerance for resource budget dimensions of different importance levels during resource allocation. For example, in the field of smart healthcare, in a hospital information system integration project, the maximum idle rate is set to 0.4 for the resource dimension of ordinary document printing equipment with a lower importance level; while for the core server resource dimension of the electronic medical record system with a higher importance level, the maximum idle rate is set to 0.15. Obtaining these maximum idle rates provides a basis for subsequent selection of target resource budget dimensions.

[0037] Subsequently, the resource idle rate of each resource budget dimension is compared with the maximum idle rate of its corresponding importance level. Resource budget dimensions whose idle rates exceed the maximum idle rate are identified and designated as target resource budget dimensions. In the smart healthcare information system integration project, it was found that the resource idle rate of some auxiliary software resource dimensions with an importance level of 3 was 0.45, while its maximum idle rate was 0.4. Therefore, this dimension was selected as a target resource budget dimension. This means that the resources in this dimension have a certain degree of idleness, exceeding the maximum idle rate allowed for its importance level, requiring resource reallocation. This screening process accurately identifies resource budget dimensions that require attention and resource reallocation, providing a clear direction for subsequent resource adjustments.

[0038] Subsequently, the resource adjustment amount corresponding to the target resource budget dimension can be calculated. Specifically, the specific resource adjustment amount can be obtained by calculating the difference between the resource idle rate and the maximum idle rate, and then combining the total resource amount or related proportional relationship of the resource budget dimension.

[0039] In actual project management, multiple projects may be running concurrently, and some projects may require additional resources for various reasons. Therefore, after calculating the resource adjustment amount, projects requiring resource replenishment can be identified from the currently executing projects. This allows resources allocated from the target resource budget to be rationally distributed to projects in need. Then, based on the urgency of the projects requiring resource replenishment, the required resource data, and the calculated resource adjustment amount, target replenishment projects are determined from the remaining projects. Here, project urgency is a crucial indicator for prioritizing resource allocation; projects with high urgency should receive priority for resource replenishment. The required resource data reflects the specific types and quantities of resources needed for each project. The resource adjustment amount corresponding to the target resource budget limits the amount of resources that can be allocated. For example, in the field of smart healthcare, in addition to information system integration projects, hospitals also have telemedicine construction projects that require additional resources. Telemedicine construction projects have certain requirements for electronic medical record system software, and their urgency is also high because they need to quickly implement telemedicine services to improve the hospital's service level. Based on the resource adjustment amount of 0.89 million yuan in the target resource budget dimension (auxiliary software), and after evaluation, it was determined that this portion of resources would be allocated to the telemedicine construction project. This process ensures that resource allocation meets the actual needs of the project while guaranteeing the rational use of resources.

[0040] Once the target supplementary projects are identified, the resource budget plan can be adjusted. Specifically, based on the resource adjustment amounts corresponding to the target resource budget dimensions, the budget amount for the corresponding dimension can be reduced from the original resource budget plan. Then, by comprehensively considering the adjusted resource budget plan, the resource adjustment amounts corresponding to the target resource budget dimensions, and the target supplementary projects, a first budget adjustment plan is generated. For example, the first budget adjustment plan might explicitly state that 0.89 million yuan is reduced from the auxiliary software budget of the information system integration project and allocated to the telemedicine construction project, presenting the adjusted resource budget status for both projects. Generating such a first budget adjustment plan provides project managers with a clear and explicit basis for resource allocation decisions, facilitating optimal resource allocation and smooth project progress.

[0041] When the matrix score is less than or equal to a preset score threshold, it indicates that the overall resource idleness of the project has not reached the standard requiring large-scale resource allocation, but there is still a certain degree of resource idleness. At this point, the amount of idle resources corresponding to each resource budget dimension can be determined based on the resource idleness matrix, and these idle resources can be marked as candidate idle resources. These candidate idle resources can serve as reserves for project resource management of the target project, and can be readily accessed during subsequent project execution should unforeseen circumstances or new resource needs arise. Then, the marked candidate idle resources and the original resource budget plan are combined to generate a first budget adjustment plan. For example, in the smart healthcare case, the first budget adjustment plan can clearly define the current resource idleness status of each resource budget dimension, mark these idle resources as candidate idle resources, and simultaneously display the overall situation of the original resource budget plan. Such a budget adjustment plan can provide project managers with detailed information on project resource utilization, helping to flexibly respond to possible resource changes during project execution and improving the flexibility and adaptability of project resource management.

[0042] Optionally, in this embodiment of the application, step 103, "based on the expected targets of each project node, calculate the predicted workload value corresponding to each project node using a preset workload prediction model," includes: for each project node, extracting node feature values ​​from the expected targets of the project node using a feature extraction agent, and inputting the node feature values ​​into a progress recognition agent, using the progress recognition agent to identify the node work progress corresponding to similar project nodes; inputting the node feature values ​​and the node work progress into the preset workload prediction model, using the preset workload prediction model to calculate the first predicted workload value corresponding to the project node based on the node feature values, and to calculate the second predicted workload value corresponding to the project node based on the node work progress, and calculating the first confidence level corresponding to the first predicted workload value and the second confidence level corresponding to the second predicted workload value, and determining the predicted workload value corresponding to the project node based on the first confidence level and the second confidence level.

[0043] In this embodiment, within the project management system, each project node has clear and specific expected goals. These goals encompass the core information of the task, such as the nature of the task, its complexity, and the types of resources required. The feature extraction agent, as an intelligent tool, utilizes natural language processing, machine learning, and other technologies to automatically and accurately extract key information from the expected goal text of each project node and transform it into node feature values. These node feature values ​​are quantitative or qualitative descriptions of the characteristics of the project node and serve as crucial foundational data for subsequent workload prediction. For example, in a payment system upgrade project in the fintech field, one project node's expected goal is "to complete the optimization of the payment interface within one month to improve the payment success rate by 5%." The feature extraction agent can extract multiple node feature values ​​from this expected goal. For instance, the task type is "payment interface optimization," which describes the essence of the task; the difficulty level can be assessed as "medium" based on experience with similar projects, reflecting the complexity of the task; and the required skills may include "payment system development" and "interface debugging," clearly defining the professional capabilities required to complete the task. By using a feature-capturing intelligent agent, key information can be extracted efficiently and accurately from the expected goals of project nodes, providing data support for subsequent workload prediction.

[0044] Subsequently, the node feature values ​​can be input into the progress recognition agent, which then performs further processing. Here, the progress recognition agent is an intelligent system built upon big data and machine learning algorithms. It possesses a vast historical project database, storing detailed information on numerous project nodes, including node feature values ​​and corresponding work progress. After inputting the node feature values ​​of the current project node into the progress recognition agent, the agent uses a similarity matching algorithm to search the historical project database for project nodes with similar feature values. Based on the found similar project nodes, it identifies the corresponding work progress of these similar project nodes. By referencing the work progress of similar project nodes, it can provide practical case studies for predicting the workload of the current project node, improving the accuracy of the prediction.

[0045] The pre-defined workload prediction model is a complex mathematical model trained and validated with a large amount of data. It can comprehensively consider the impact of multiple factors on the workload of project nodes. When node feature values ​​and node work progress are input into the model, the model can perform calculations from two different perspectives.

[0046] Firstly, the model can calculate the initial workload prediction based on node feature values. Specifically, the model can use built-in algorithms and weights to comprehensively analyze factors such as task type, difficulty level, and required skills. For example, for the payment interface optimization node in a fintech project, the model can calculate a reasonable workload prediction, assuming it to be 25 working days, based on factors such as the common workload range for the task type "payment interface optimization," the additional workload bonus corresponding to the "medium" difficulty level, and the proficiency requirements for the required skills "payment system development" and "interface debugging."

[0047] Secondly, a second workload prediction can be calculated based on the progress of node work. Specifically, the model can refer to the work progress of nodes in similar projects and adjust it according to the actual situation of the current project. Taking a fintech project as an example, the model referenced the situation where a node in a similar project completed the optimization work in 28 days. Considering the differences between the current project and similar projects in terms of resource allocation and technical difficulty, the calculated second workload prediction after adjustment can be 27 working days.

[0048] Using the two calculation methods described above, the preset workload prediction model can predict the workload of project nodes from different perspectives, providing multiple references for subsequent decision-making.

[0049] Confidence level is an important indicator for measuring the reliability of a prediction, reflecting the degree to which the predicted value closely approximates the actual value. Calculating the confidence levels for the first and second workload predictions can help assess the accuracy and reliability of these two predictions. The method for calculating confidence levels can be based on the characteristics of a pre-defined workload prediction model and historical data. For the first workload prediction, factors such as model stability (consistent performance across different datasets), data accuracy (accurate extraction of node feature values, quality of historical data), and the deviation between the prediction and the actual results can be considered. For example, in a fintech project, for a first workload prediction of 25 working days, by analyzing model stability (small prediction errors across multiple similar projects), data accuracy (accurate node feature value extraction), and the deviation between historical similar predictions and the actual results (within a reasonable range), the calculated first confidence level is 0.85. For the second workload prediction, in addition to considering the above factors, the similarity between similar project nodes and the current project can also be emphasized. If similar project nodes are highly similar to the current project in terms of task type, difficulty, and resources, then the confidence level of the second workload forecast will be relatively high. In fintech projects, for a second workload forecast of 27 working days, considering the high similarity between similar project nodes and the current project, the calculated second confidence level is 0.8. Calculating the confidence level provides a more comprehensive understanding of the reliability of the two workload forecasts, offering a basis for determining the final workload forecast.

[0050] Subsequently, based on the first and second confidence levels, various rules or algorithms can be used to determine the predicted workload for each project node. For example, both confidence levels can be considered together, and the predicted value with the higher confidence level can be selected as the final predicted workload. Alternatively, a weighted average method can be used, assigning different weights to the two predicted values ​​according to their respective confidence levels, and then calculating the weighted average as the final predicted workload. By reasonably determining the predicted workload, a scientific basis can be provided for resource allocation and schedule planning of the target project, thereby improving the efficiency and success rate of project management.

[0051] Optionally, in this embodiment of the application, step 103, "dynamically determining the second budget adjustment scheme corresponding to the resource budget scheme based on the predicted workload value corresponding to each project node," includes: determining the project nodes covered by the resource budget scheme based on the execution order corresponding to each project node, the predicted workload value corresponding to each project node, and the resource budget scheme; adjusting the resource budget scheme based on the project nodes covered by the resource budget scheme to obtain the second budget adjustment scheme; and marking the target project as a resource replenishment project.

[0052] In this embodiment, the execution order of project nodes represents the timeline of project progress, determining the sequence in which tasks are carried out. The predicted workload for each project node reflects the resources required to complete its tasks. Finally, by combining the execution order of project nodes in the target project, the predicted workload for each node, and the initial resource budget plan, the project nodes covered by the resource budget plan can be determined. In other words, based on the project's progress order and the workload requirements of each node, project nodes that can be supported by current resources are selected within the resource budget framework. For example, in a fintech payment system upgrade project, the available human, material, and time resources for development within the resource budget plan are limited. Following the execution order, the payment interface optimization node is considered first. If its predicted workload is within the resource budget's capacity, then this node is included in the project nodes covered by the resource budget plan. Next, the security module upgrade node is considered. If its predicted workload plus the total workload of already covered nodes does not exceed the resource budget, it is also included, and so on, until the resource budget can no longer support the workload requirements of the next node.

[0053] Once the project milestones covered by the resource budget plan are identified, the plan can be fine-tuned based on the actual conditions of these milestones to ensure the overall project schedule remains unaffected. This adjustment results in a second budget plan that more accurately matches actual project needs and allows for timely identification of which project milestones are experiencing resource shortages. Furthermore, the target project can be marked as a resource-replenishing project. This allows for the search for additional resources to ensure the project's smooth progress. Marking a target project as a resource-replenishing project provides project managers and stakeholders with a clear understanding of the project's resource status, enabling timely measures to address resource shortages and ensuring the project's successful completion as planned. Simultaneously, when other projects have available resources, these resources can be automatically allocated to the target project, achieving flexible resource allocation and efficient utilization.

[0054] Furthermore, as Figure 1 To specifically implement the method, this application provides a resource allocation device based on dynamic control, such as... Figure 2 As shown, the device includes: The prediction module is used to obtain the resource budget plan and project execution data corresponding to the target project, and calculate the predicted resource requirements of the target project based on the project execution data through a preset project resource prediction model. The first adjustment module is used to calculate a resource idle matrix based on the resource budget plan and the predicted resource demand if the resource budget plan covers the predicted resource demand, and to calculate the matrix score corresponding to the resource idle matrix. Based on the matrix score, the module dynamically determines a first budget adjustment plan corresponding to the resource budget plan, so as to allocate resources based on the first budget adjustment plan. The second adjustment module is used to obtain multiple project nodes corresponding to the target project and the expected target of each project node if the resource budget plan does not cover the predicted resource demand. Based on the expected target of each project node, the module calculates the predicted workload value of each project node through a preset workload prediction model. Based on the predicted workload value of each project node, the module dynamically determines the second budget adjustment plan corresponding to the resource budget plan, and allocates resources based on the second budget adjustment plan.

[0055] Optionally, the preset project resource prediction model performs the following operations to obtain the predicted resource requirements of the target project: Extract structured and unstructured data from the project execution data; Based on the structured data, the project characteristic index data corresponding to the target project is determined, and the project characteristic index data is input into the time series analysis sub-model and the regression analysis sub-model respectively. The first predicted demand is obtained through the time series analysis sub-model, and the second predicted demand is obtained through the regression analysis sub-model. Data mining is performed on the unstructured data to obtain supplementary project indicator data, and the supplementary project indicator data is input into the demand prediction agent to obtain the third predicted demand. Based on the first forecasted demand, the second forecasted demand, and the third forecasted demand, the forecasted resource requirements of the target project are determined.

[0056] Optionally, the first adjustment module is used to: Based on the resource budget dimensions included in the resource budget scheme, determine the number of columns corresponding to the resource idle matrix; For each resource budget dimension, the resource availability corresponding to the resource budget dimension is calculated using the resource budget scheme and the predicted resource demand, and the resource availability corresponding to each resource budget dimension is mapped to the first row of the resource availability matrix; For each resource budget dimension, the importance level corresponding to the resource budget dimension is determined by a preset dimension importance level mapping table, and the importance level corresponding to each resource budget dimension is mapped to the second row of the resource idle matrix.

[0057] Optionally, the first adjustment module is further configured to: The resource idle amount under each resource budget dimension in the resource idle matrix is ​​scaled proportionally to obtain the resource idle rate corresponding to each resource idle amount. For each resource budget dimension in the resource idle matrix, the resource idle rate and importance level under the resource budget dimension are multiplied to obtain the product result under the resource budget dimension. The product results under each resource budget dimension are summed to obtain the matrix score corresponding to the resource idle matrix.

[0058] Optionally, the first adjustment module is further configured to: If the matrix score is greater than a preset score threshold, the maximum idle rate corresponding to each importance level is obtained. Target resource budget dimensions with resource idle rates greater than the corresponding maximum idle rates are selected from the resource idle matrix. Based on the resource idle rate and maximum idle rate corresponding to the target resource budget dimension, the resource adjustment amount corresponding to the target resource budget dimension is calculated. From the currently executing projects, resource-to-be-added projects are obtained, and based on the project urgency, resource-to-be-added data, and resource adjustment amount corresponding to the target resource budget dimension, the target supplementary project corresponding to the resource adjustment amount is determined. The resource budget plan is adjusted according to the resource adjustment amount corresponding to the target resource budget dimension, and a first budget adjustment plan is generated based on the adjusted resource budget plan, the resource adjustment amount corresponding to the target resource budget dimension, and the target supplementary project. If the matrix score is less than or equal to the preset score threshold, then based on the resource idle matrix, the resource idle amount corresponding to each resource budget dimension is determined, the resource idle amount is marked as a candidate idle resource, and a first budget adjustment plan is generated based on the marked candidate idle resources and the resource budget plan.

[0059] Optionally, the second adjustment module is used for: For each project node, a feature extraction agent extracts node feature values ​​from the expected targets of the corresponding project node, and inputs the node feature values ​​into a progress recognition agent, which then identifies the work progress of nodes corresponding to similar project nodes. The node feature values ​​and the node work progress are input into the preset workload prediction model. The preset workload prediction model calculates the first workload prediction value corresponding to the project node based on the node feature values ​​and the second workload prediction value corresponding to the project node based on the node work progress. The first confidence level corresponding to the first workload prediction value and the second confidence level corresponding to the second workload prediction value are calculated respectively. The workload prediction value corresponding to the project node is determined based on the first confidence level and the second confidence level.

[0060] Optionally, the second adjustment module is further configured to: Based on the execution order of each project node, the predicted workload of each project node, and the resource budget plan, the project nodes covered by the resource budget plan are determined. The resource budget plan is then adjusted based on the project nodes covered by the resource budget plan to obtain a second budget adjustment plan, and the target project is marked as a project to be supplemented with resources.

[0061] It should be noted that other corresponding descriptions of the functional units involved in the resource allocation device based on dynamic control provided in this application embodiment can be found by referring to... Figure 1 The corresponding descriptions in the method will not be repeated here.

[0062] This application also provides a computer device, which may specifically be a personal computer, a server, a network device, etc. Figure 3 As shown, the computer device includes a bus, a processor, memory, and a communication interface, and may also include an input / output interface and a display device. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores location information. The network interface allows communication with external terminals via a network connection. When the computer program is executed by the processor, it implements the steps in the various method embodiments.

[0063] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0064] In one embodiment, a computer-readable storage medium is provided, which may be non-volatile or volatile, having stored thereon a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0065] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0066] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0067] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0068] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0069] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A resource allocation method based on dynamic control, characterized in that, include: Obtain the resource budget plan and project execution data corresponding to the target project, and calculate the predicted resource requirements of the target project based on the project execution data using a preset project resource prediction model; If the resource budget plan covers the predicted resource demand, then a resource idle matrix is ​​calculated based on the resource budget plan and the predicted resource demand, and a matrix score corresponding to the resource idle matrix is ​​calculated. A first budget adjustment plan corresponding to the resource budget plan is dynamically determined based on the matrix score, so as to allocate resources based on the first budget adjustment plan. If the resource budget plan does not cover the predicted resource demand, then obtain multiple project nodes corresponding to the target project, and the node expected target corresponding to each project node. Based on the node expected target corresponding to each project node, calculate the predicted workload value corresponding to each project node through a preset workload prediction model. Based on the predicted workload value corresponding to each project node, dynamically determine the second budget adjustment plan corresponding to the resource budget plan, and allocate resources based on the second budget adjustment plan.

2. The method according to claim 1, characterized in that, The preset project resource prediction model performs the following operations to obtain the predicted resource requirements of the target project: Extract structured and unstructured data from the project execution data; Based on the structured data, the project characteristic index data corresponding to the target project is determined, and the project characteristic index data is input into the time series analysis sub-model and the regression analysis sub-model respectively. The first predicted demand is obtained through the time series analysis sub-model, and the second predicted demand is obtained through the regression analysis sub-model. Data mining is performed on the unstructured data to obtain supplementary project indicator data, and the supplementary project indicator data is input into the demand prediction agent to obtain the third predicted demand. Based on the first forecasted demand, the second forecasted demand, and the third forecasted demand, the forecasted resource requirements of the target project are determined.

3. The method according to claim 1, characterized in that, The step of calculating the resource availability matrix based on the resource budget scheme and the predicted resource demand includes: Based on the resource budget dimensions included in the resource budget scheme, determine the number of columns corresponding to the resource idle matrix; For each resource budget dimension, the resource availability corresponding to the resource budget dimension is calculated using the resource budget scheme and the predicted resource demand, and the resource availability corresponding to each resource budget dimension is mapped to the first row of the resource availability matrix; For each resource budget dimension, the importance level corresponding to the resource budget dimension is determined by a preset dimension importance level mapping table, and the importance level corresponding to each resource budget dimension is mapped to the second row of the resource idle matrix.

4. The method according to claim 3, characterized in that, The calculation of the matrix score corresponding to the resource idle matrix includes: The resource idle amount under each resource budget dimension in the resource idle matrix is ​​scaled proportionally to obtain the resource idle rate corresponding to each resource idle amount. For each resource budget dimension in the resource idle matrix, the resource idle rate and importance level under the resource budget dimension are multiplied to obtain the product result under the resource budget dimension. The product results under each resource budget dimension are summed to obtain the matrix score corresponding to the resource idle matrix.

5. The method according to claim 4, characterized in that, The step of dynamically determining the first budget adjustment plan corresponding to the resource budget plan based on the matrix scoring includes: If the matrix score is greater than a preset score threshold, the maximum idle rate corresponding to each importance level is obtained. Target resource budget dimensions with resource idle rates greater than the corresponding maximum idle rates are selected from the resource idle matrix. Based on the resource idle rate and maximum idle rate corresponding to the target resource budget dimension, the resource adjustment amount corresponding to the target resource budget dimension is calculated. From the currently executing projects, resource-to-be-added projects are obtained, and based on the project urgency, resource-to-be-added data, and resource adjustment amount corresponding to the target resource budget dimension, the target supplementary project corresponding to the resource adjustment amount is determined. The resource budget plan is adjusted according to the resource adjustment amount corresponding to the target resource budget dimension, and a first budget adjustment plan is generated based on the adjusted resource budget plan, the resource adjustment amount corresponding to the target resource budget dimension, and the target supplementary project. If the matrix score is less than or equal to the preset score threshold, then based on the resource idle matrix, the resource idle amount corresponding to each resource budget dimension is determined, the resource idle amount is marked as a candidate idle resource, and a first budget adjustment plan is generated based on the marked candidate idle resources and the resource budget plan.

6. The method according to claim 1, characterized in that, Based on the expected targets for each project node, and using a pre-set workload prediction model, the predicted workload value for each project node is calculated, including: For each project node, a feature extraction agent extracts node feature values ​​from the expected targets of the corresponding project node, and inputs the node feature values ​​into a progress recognition agent, which then identifies the work progress of nodes corresponding to similar project nodes. The node feature values ​​and the node work progress are input into the preset workload prediction model. The preset workload prediction model calculates the first workload prediction value corresponding to the project node based on the node feature values ​​and the second workload prediction value corresponding to the project node based on the node work progress. The first confidence level corresponding to the first workload prediction value and the second confidence level corresponding to the second workload prediction value are calculated respectively. The workload prediction value corresponding to the project node is determined based on the first confidence level and the second confidence level.

7. The method according to claim 1, characterized in that, The step of dynamically determining the second budget adjustment plan corresponding to the resource budget plan based on the predicted workload values ​​corresponding to each project node includes: Based on the execution order of each project node, the predicted workload of each project node, and the resource budget plan, the project nodes covered by the resource budget plan are determined. The resource budget plan is then adjusted based on the project nodes covered by the resource budget plan to obtain a second budget adjustment plan, and the target project is marked as a project to be supplemented with resources.

8. A resource allocation device based on dynamic control, characterized in that, include: The prediction module is used to obtain the resource budget plan and project execution data corresponding to the target project, and calculate the predicted resource requirements of the target project based on the project execution data through a preset project resource prediction model. The first adjustment module is used to calculate a resource idle matrix based on the resource budget plan and the predicted resource demand if the resource budget plan covers the predicted resource demand, and to calculate the matrix score corresponding to the resource idle matrix. Based on the matrix score, the module dynamically determines a first budget adjustment plan corresponding to the resource budget plan, so as to allocate resources based on the first budget adjustment plan. The second adjustment module is used to obtain multiple project nodes corresponding to the target project and the expected target of each project node if the resource budget plan does not cover the predicted resource demand. Based on the expected target of each project node, the module calculates the predicted workload value of each project node through a preset workload prediction model. Based on the predicted workload value of each project node, the module dynamically determines the second budget adjustment plan corresponding to the resource budget plan, and allocates resources based on the second budget adjustment plan.

9. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

10. A computer device, comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 7.

Images