CN-122001955-A - Resource scheduling method, device and equipment

CN122001955ACN 122001955 ACN122001955 ACN 122001955ACN-122001955-A

Abstract

The embodiment of the invention discloses a resource scheduling method, a device and equipment, wherein the method comprises the steps of receiving a resource scheduling request for a target communication item to be built, responding to the resource scheduling request, determining a to-be-scheduled area corresponding to the target communication item, acquiring schedulable resource information of the to-be-scheduled area, determining a plurality of candidate scheduling strategies and action values corresponding to each candidate scheduling strategy based on the schedulable resource information by utilizing a pre-trained decision model based on the importance weight of the to-be-scheduled area, adjusting the action values corresponding to the candidate scheduling strategies based on the importance weight of the to-be-scheduled area, determining a target scheduling strategy from the candidate scheduling strategies based on the adjusted action values, and performing resource scheduling processing on the target communication item based on the target scheduling strategy.

Inventors

ZHENG HAN
YAO LI
HUANG KEJI
WENG CHENHUAN
XIE YAQIAN

Assignees

中国移动通信集团浙江有限公司
中国移动通信集团有限公司
中国移动(浙江)创新研究院有限公司
浙江移动数智科技有限公司

Dates

Publication Date: 20260508
Application Date: 20251223

Claims (10)

1. A method for scheduling resources, the method comprising: receiving a resource scheduling request for a target communication item to be constructed; Responding to the resource scheduling request, determining a to-be-scheduled area corresponding to the target communication item, acquiring schedulable resource information of the to-be-scheduled area, and determining importance weight based on the resource quantity and/or the position information of the to-be-scheduled area; determining a plurality of candidate scheduling strategies and action values corresponding to each candidate scheduling strategy based on schedulable resource information by utilizing a pre-trained decision model, and adjusting the action values corresponding to the candidate scheduling strategies based on importance weights of the areas to be scheduled; Determining a target scheduling strategy from the candidate scheduling strategies based on the adjusted action value; and carrying out resource scheduling processing on the target communication item based on the target scheduling policy.
2. The method of claim 1, wherein determining a target scheduling policy from the candidate scheduling policies based on the adjusted action value comprises: Determining a first scheduling policy from the candidate scheduling policies based on the adjusted action value; Determining a target scheduling resource corresponding to the first scheduling policy, and acquiring working efficiency of the target scheduling resource and influence evaluation data of the target scheduling resource on an area to be scheduled after the target scheduling resource is scheduled; Determining a reward value corresponding to the first resource scheduling policy based on the engineering quantity of the target communication project, the working efficiency of the target scheduling resource and the influence evaluation data; and screening the scheduling strategies meeting the construction requirements of the target communication project from the candidate scheduling strategies based on the rewarding values corresponding to the first resource scheduling strategies, and determining the screened scheduling strategies as the target scheduling strategies.
3. The method of claim 2, wherein the determining, using a pre-trained decision model, a plurality of candidate scheduling policies and action values corresponding to each of the candidate scheduling policies based on schedulable resource information, and adjusting the action values corresponding to the candidate scheduling policies based on importance weights of the areas to be scheduled, comprises: constructing graph structure data corresponding to the to-be-scheduled area based on the resource state and the importance weight of the to-be-scheduled area, wherein the graph structure data comprises nodes corresponding to the resource state of the to-be-scheduled area and edges used for representing the importance weight of the to-be-scheduled area; determining the plurality of candidate scheduling strategies and action values corresponding to each candidate scheduling strategy based on schedulable resource information by utilizing a strategy network of the pre-trained decision model; And determining a node embedded vector corresponding to the graph structure data by utilizing an extraction network of the pre-trained decision model, and adjusting the action value corresponding to the candidate scheduling strategy based on the node embedded vector.
4. The method of claim 3, further comprising, prior to the determining the plurality of candidate scheduling policies and the action value corresponding to each of the candidate scheduling policies based on schedulable resource information by the policy network utilizing the pre-trained decision model: Acquiring training data for training the decision model, wherein the training data comprises historical schedulable resource information and importance weights of historical scheduling areas; Determining a predictive scheduling strategy and a first action value corresponding to the predictive scheduling strategy based on the historical schedulable resource information and the importance weight of the historical scheduling area by utilizing a strategy network of a decision model; determining a second action value corresponding to the predictive scheduling strategy by using a target network of the decision model, wherein the target network has the same network structure as the strategy network, and the target network copies network parameters from the strategy network based on a preset updating period; determining a loss value based on the first action value and the second action value, and updating network parameters of the strategy network based on the loss value; Judging whether the updated decision model meets a preset convergence condition, and continuously training the decision model based on the training data under the condition that the updated decision model does not meet the preset convergence condition until the decision model converges to obtain a trained decision model.
5. The method of claim 2, wherein the determining the prize value corresponding to the first resource scheduling policy based on the engineering quantity of the target communication item, the operational effectiveness of the target scheduling resource, and the impact assessment data comprises: Determining an estimated completion time based on the engineering quantity of the target communication project and the work efficiency of the target scheduling resource; and acquiring the set completion time of the target communication project, and determining a reward value corresponding to the first resource scheduling strategy based on the ratio of the expected completion time to the set completion time and the influence evaluation data.
6. The method of claim 5, wherein the determining a prize value for the first resource scheduling policy based on the ratio of the predicted completion time to the set completion time and the impact assessment data comprises: Determining a reward value corresponding to the first resource scheduling policy based on a preset positive reward value, a ratio of the predicted completion time to the set completion time, and the impact evaluation data, if the predicted completion time is not greater than the set completion time; and under the condition that the predicted completion time is larger than the set completion time, determining the reward value corresponding to the first resource scheduling strategy based on a preset negative reward value and the ratio of the predicted completion time to the set completion time.
7. A resource scheduling apparatus, the apparatus comprising: The request receiving module is used for receiving a resource scheduling request aiming at a target communication project to be built; The data acquisition module is used for responding to the resource scheduling request, determining a to-be-scheduled area corresponding to the target communication item, acquiring schedulable resource information of the to-be-scheduled area and determining importance weights based on the resource quantity and/or the position information of the to-be-scheduled area; The value adjustment module is used for determining a plurality of candidate scheduling strategies and action values corresponding to each candidate scheduling strategy based on schedulable resource information by utilizing a pre-trained decision model, and adjusting the action values corresponding to the candidate scheduling strategies based on importance weights of the areas to be scheduled; The strategy determining module is used for determining a target scheduling strategy from the candidate scheduling strategies based on the adjusted action value; And the resource scheduling module is used for carrying out resource scheduling processing on the target communication item based on the target scheduling strategy.
8. A resource scheduling device comprising a processor, a memory and a computer program stored on the memory and executable on the processor, which when executed by the processor implements the steps of the resource scheduling method of any one of claims 1 to 6.
9. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of the resource scheduling method according to any of claims 1 to 6.
10. A computer program product comprising a computer program which, when executed by a processor, implements the steps of the resource scheduling method of any one of claims 1 to 6.

Description

Resource scheduling method, device and equipment Technical Field The present invention relates to the field of computer technologies, and in particular, to a method, an apparatus, and a device for scheduling resources. Background In the process of communication network construction and maintenance, it is often necessary to schedule communication security personnel across regions to cope with communication projects such as large engineering projects, emergency repair tasks or regional network upgrades. The conventional scheduling method mainly depends on manual experience, but as the communication project to be built is more and more complex, the manual experience is difficult to handle the problem of high-dimensional and nonlinear complex scheduling, so that the resource scheduling efficiency and the collaborative optimization efficiency for the communication project are low, and therefore, a technical scheme capable of improving the resource scheduling efficiency and the global collaborative optimization effect for the communication project to be built is needed. Disclosure of Invention The embodiment of the invention aims to provide a technical scheme capable of improving resource scheduling efficiency and global collaborative optimization effect aiming at communication projects to be built. In order to solve the technical problems, the embodiment of the invention is realized as follows: In a first aspect, a method for scheduling resources provided by an embodiment of the present invention includes: The method comprises the steps of receiving a resource scheduling request for a target communication project to be built, responding to the resource scheduling request, determining a to-be-scheduled area corresponding to the target communication project, obtaining schedulable resource information of the to-be-scheduled area and importance weights determined based on the resource quantity and/or the position information of the to-be-scheduled area, determining a plurality of candidate scheduling strategies and action values corresponding to each candidate scheduling strategy based on the schedulable resource information by utilizing a pre-trained decision model, adjusting the action values corresponding to the candidate scheduling strategies based on the importance weights of the to-be-scheduled area, determining a target scheduling strategy from the candidate scheduling strategies based on the adjusted action values, and carrying out resource scheduling processing on the target communication project based on the target scheduling strategy. In a second aspect, an embodiment of the present invention provides a resource scheduling apparatus, where the apparatus includes: The resource scheduling system comprises a request receiving module, a data acquisition module, a strategy determination module and a resource scheduling module, wherein the request receiving module is used for receiving a resource scheduling request for a target communication project to be built, the data acquisition module is used for responding to the resource scheduling request, determining a to-be-scheduled area corresponding to the target communication project, acquiring schedulable resource information of the to-be-scheduled area and importance weights determined based on the number and/or position information of resources of the to-be-scheduled area, the value adjustment module is used for determining a plurality of candidate scheduling strategies and action values corresponding to each candidate scheduling strategy based on the schedulable resource information by utilizing a pre-trained decision model, and adjusting the action values corresponding to the candidate scheduling strategies based on the importance weights of the to-be-scheduled area, the strategy determination module is used for determining a target scheduling strategy from the candidate scheduling strategies based on the adjusted action values, and the resource scheduling module is used for carrying out resource scheduling processing on the target communication project. In a third aspect, an embodiment of the present invention provides a resource scheduling device, including a processor, a memory, and a computer program stored on the memory and executable on the processor, where the computer program when executed by the processor implements the steps of the resource scheduling method provided in the foregoing embodiment. In a fourth aspect, an embodiment of the present invention provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements the steps of the resource scheduling method provided in the above embodiment. In a fifth aspect, an embodiment of the present invention provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the resource scheduling method provided by the above embodiment. Drawings In order to more clearly illustrate t