Search

CN-121981459-A - Resource quantity early warning method, device, equipment, medium and product for research and development project

CN121981459ACN 121981459 ACN121981459 ACN 121981459ACN-121981459-A

Abstract

The invention discloses a resource quantity early warning method, device, equipment, medium and product of a research and development project, which relates to the technical field of artificial intelligence and can be applied to the technical field of finance and technology, and comprises the steps of obtaining historical attribute data of a target research and development project, and predicting the resource demand of the target research and development project at the current moment according to a moment influence factor and the historical attribute data at the current moment; determining a resource quantity entropy value of a target research and development project at the current moment according to the resource remaining quantity and the resource demand quantity of the target system at the current moment, determining a dynamic resource quantity entropy threshold corresponding to the current moment according to the time difference between the threshold attenuation moment and the current moment, and carrying out resource quantity early warning on the target research and development project according to the resource quantity entropy value and the dynamic resource quantity entropy threshold. The method reduces the error rate of resource demand prediction, improves the accuracy of resource demand early warning, and reduces the problems of early warning false report or early warning missing.

Inventors

  • XU JIA

Assignees

  • 中国工商银行股份有限公司

Dates

Publication Date
20260505
Application Date
20260116

Claims (13)

  1. 1. The resource quantity early warning method for the research and development project is characterized by comprising the following steps of: Acquiring historical attribute data of a target research and development project in a historical time period, and predicting the resource demand of the target research and development project at the current time according to a time influence factor corresponding to the current time and the historical attribute data, wherein the historical attribute data comprises at least one of historical resource usage and historical code verification rate; Determining a resource quantity entropy value of a target research and development project at the current moment according to the resource residual quantity and the resource demand quantity of the target system at the current moment, and determining a dynamic resource quantity entropy threshold corresponding to the current moment according to a time difference value between a threshold attenuation moment and the current moment, wherein the target system is a system to which the target research and development project belongs; And carrying out resource quantity early warning on the target research and development project according to the resource quantity entropy value and the dynamic resource quantity entropy threshold value.
  2. 2. The method according to claim 1, wherein the determining the dynamic resource amount entropy threshold corresponding to the current time according to the time difference between the threshold decay time and the current time comprises: Determining a maximum threshold attenuation value according to the difference value between the maximum resource amount entropy threshold and the minimum resource amount entropy threshold, and determining a threshold attenuation proportional coefficient according to the product result between the current threshold attenuation rate and the time difference value; and determining a threshold attenuation value according to a product result between the maximum threshold attenuation value and the threshold attenuation proportion coefficient, and determining the dynamic resource amount entropy threshold according to a difference value between the maximum resource amount entropy threshold and the threshold attenuation value.
  3. 3. The method of claim 2, wherein the current threshold decay rate is determined by: determining a resource quantity index and a task execution index of the target system at the current moment, and determining a historical feedback index obtained by carrying out resource quantity early warning on the target research and development project based on a historical resource quantity entropy threshold, wherein the historical resource quantity entropy threshold is determined according to a historical threshold attenuation rate; Generating state description information according to the resource quantity index, the task execution index, the historical feedback index and the historical threshold attenuation rate, and inputting the state description information into a threshold attenuation rate adjustment model; and determining a threshold attenuation rate adjustment value according to the state description information through the threshold attenuation rate adjustment model, and determining the current threshold attenuation rate according to the historical threshold attenuation rate and the threshold attenuation rate adjustment value.
  4. 4. The method of claim 3, wherein the resource amount indicator comprises at least one of a CPU usage amount, a memory usage amount, and a bandwidth usage amount, the task execution indicator comprises at least one of a task queue length, a task execution average time consumption, and a task execution error rate, and the historical feedback indicator comprises at least one of an early warning false alarm rate and an early warning false alarm rate.
  5. 5. The method according to claim 3, further comprising, after the resource amount pre-warning for the target development project: determining the number of early warning false alarms, the number of early warning missing alarms and the running stability index of the target system, and determining a threshold attenuation rate adjustment score according to the number of early warning false alarms, the number of early warning missing alarms and the running stability index; generating a state value according to the state description information, generating an action value according to the threshold decay rate adjustment value, generating a reward value according to the threshold decay rate adjustment score, and performing reinforcement learning on the threshold decay rate adjustment model according to the state value, the action value and the reward value.
  6. 6. The method according to claim 1, wherein predicting the resource demand of the target development project at the current time according to the time impact factor corresponding to the current time and the history attribute data comprises: According to the time influencing factors and the historical attribute data, predicting the resource demand by adopting a first prediction method to obtain a first predicted demand, and according to the time influencing factors and the historical attribute data, predicting the resource demand by adopting a second prediction method to obtain a second predicted demand; determining a first prediction weight corresponding to the first prediction method, determining a second prediction weight corresponding to the second prediction method, and performing weighted calculation according to the first prediction demand, the second prediction demand, the first prediction weight and the second prediction weight to obtain the resource demand.
  7. 7. The method according to claim 1, wherein determining the entropy value of the resource amount of the target development project at the current time according to the resource remaining amount at the current time and the resource demand amount comprises: performing difference operation according to the resource demand and the resource residual quantity to determine a resource gap quantity; and carrying out ratio operation according to the resource gap amount and the total resource amount of the target system to obtain the resource amount entropy value.
  8. 8. The method of claim 1, further comprising, after the resource amount pre-warning for the target development project: Under the condition that the available resource quantity of the target research and development project is insufficient, determining the idle resource quantity in an idle resource pool, and determining the scheduling resource quantity according to the size relation among the target resource quantity, the resource gap quantity and the idle resource quantity, wherein the target resource quantity is determined according to the product result between the total resource quantity of the target system and the preset proportion; And scheduling target resources from the idle resource pool according to the scheduling resource quantity, and distributing the target resources to the target research and development project.
  9. 9. The method of claim 8, wherein the determining the amount of scheduled resources based on the size relationship between the target amount of resources, the amount of resource gaps, and the amount of free resources comprises: and determining a minimum value among the target resource quantity, the resource gap quantity and the idle resource quantity, and determining the scheduling resource quantity according to the minimum value.
  10. 10. A resource amount early warning device for a research and development project, the device comprising: The resource demand prediction module is used for acquiring historical attribute data of a target research and development project in a historical time period, and predicting the resource demand of the target research and development project at the current time according to a time influence factor corresponding to the current time and the historical attribute data, wherein the historical attribute data comprises at least one of historical resource usage and historical code verification rate; The dynamic resource amount entropy threshold determining module is used for determining a resource amount entropy value of the target research and development project at the current moment according to the resource remaining amount of the target system at the current moment and the resource demand, and determining a dynamic resource amount entropy threshold corresponding to the current moment according to a time difference value between a threshold attenuation moment and the current moment, wherein the target system is a system to which the target research and development project belongs; and the resource quantity early warning module is used for carrying out resource quantity early warning on the target research and development project according to the resource quantity entropy value and the dynamic resource quantity entropy threshold value.
  11. 11. An electronic device, the electronic device comprising: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-9.
  12. 12. A computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, the computer instructions for causing a processor to perform the method of any one of claims 1-9.
  13. 13. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any of claims 1-9.

Description

Resource quantity early warning method, device, equipment, medium and product for research and development project Technical Field The invention relates to the technical field of artificial intelligence, and can be applied to the field of financial science and technology, in particular to a resource quantity early warning method, device, equipment, medium and product for research and development projects. Background The full-period resource management of the research and development project is a core link for guaranteeing the technological achievement to fall to the ground. As the complexity of research and development increases, the dynamics of resource consumption becomes a key challenge for project management. In the resource management of research and development projects, the monitoring and early warning of the resource quantity are key to guaranteeing the on-schedule delivery of the projects. In the existing research and development project resource management, the resource quantity required by the research and development project is usually predicted by means of manual experience, and the resource quantity early warning is carried out by setting a fixed threshold value, so that the problems of inaccurate resource demand quantity prediction and resource quantity early warning exist. Disclosure of Invention The invention provides a resource quantity early warning method, device, equipment, medium and product of a research and development project, which are used for solving the problems of inaccurate resource demand quantity prediction and resource quantity early warning in the existing research and development project resource management. According to one aspect of the invention, there is provided a resource amount early warning method for a research and development project, the method comprising: Acquiring historical attribute data of a target research and development project in a historical time period, and predicting the resource demand of the target research and development project at the current time according to a time influence factor corresponding to the current time and the historical attribute data, wherein the historical attribute data comprises at least one of historical resource usage and historical code verification rate; Determining a resource quantity entropy value of a target research and development project at the current moment according to the resource residual quantity and the resource demand quantity of the target system at the current moment, and determining a dynamic resource quantity entropy threshold corresponding to the current moment according to a time difference value between a threshold attenuation moment and the current moment, wherein the target system is a system to which the target research and development project belongs; And carrying out resource quantity early warning on the target research and development project according to the resource quantity entropy value and the dynamic resource quantity entropy threshold value. According to another aspect of the present invention, there is provided a resource amount early warning device for a development project, the device comprising: The resource demand prediction module is used for acquiring historical attribute data of a target research and development project in a historical time period, and predicting the resource demand of the target research and development project at the current time according to a time influence factor corresponding to the current time and the historical attribute data, wherein the historical attribute data comprises at least one of historical resource usage and historical code verification rate; The dynamic resource amount entropy threshold determining module is used for determining a resource amount entropy value of the target research and development project at the current moment according to the resource remaining amount of the target system at the current moment and the resource demand, and determining a dynamic resource amount entropy threshold corresponding to the current moment according to a time difference value between a threshold attenuation moment and the current moment, wherein the target system is a system to which the target research and development project belongs; and the resource quantity early warning module is used for carrying out resource quantity early warning on the target research and development project according to the resource quantity entropy value and the dynamic resource quantity entropy threshold value. According to another aspect of the present invention, there is provided an electronic apparatus including: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of any one of the present invention. According to another aspect of the invention there is provided a computer readable stor