Artificial intelligence task scheduling method based on deep learning

Abstract

The invention belongs to the technical field of task scheduling, and provides an artificial intelligence task scheduling method based on deep learning, which comprises the following steps: analyzing and processing a task by using a deep learning algorithm, performing deep learning and processing on the task based on performance data to obtain a task evaluation value, grading the task according to the task evaluation value, and performing task scheduling. The method comprises the following steps: acquiring task data based on comparison of a task evaluation value and a task evaluation standard value, acquiring a task completion value and an advanced task completion value based on the task data, processing to obtain an advanced performance value, comparing the advanced performance value with an advanced performance threshold value, marking a worker according to a comparison result, and marking the worker according to the comparison of the advanced performance value. And comparing the number of the high-level tasks with the number of the high-level personnel to obtain an optimized scheduling value, carrying out combined analysis on the optimized scheduling value YH and the high-level performance value BX to obtain an optimized scheduling coefficient, redistributing the high-level tasks according to the optimized scheduling coefficient, and establishing a redistribution task library.

Description

Technical Field

[0001] The present invention belongs to the technical field of task scheduling, and specifically is an artificial intelligence task scheduling method based on deep learning. Background Art

[0002] With the rapid development of artificial intelligence technology, deep learning models have been widely used in various fields. However, the training and reasoning processes of deep learning models usually require a lot of computing resources. How to efficiently and reasonably allocate these resources to meet the needs of different tasks has become an urgent problem to be solved. Traditional task scheduling methods are often based on pre-set rules or static parameters, which are difficult to adapt to dynamically changing resource environments and task requirements. Therefore, designing a task scheduling method based on deep learning that can adaptively adjust resource allocation has important practical significance and application value.

[0003] A Chinese patent application with publication number CN116542491A discloses an artificial intelligence task scheduling method based on deep learning, including: a first data acquisition module, a first-level marking module, a second data acquisition module, a second-level marking module, and a screening module; the first data acquisition module collects historical record information of emergency repair personnel repairing power failures of the same nature and generates an emergency repair evaluation index; the first-level marking module compares the acquired emergency repair evaluation index with the emergency repair evaluation index reference threshold, marks the emergency repair personnel as highly selectable personnel and low selectable personnel, and passes the high selectable personnel mark and the low selectable personnel mark to the screening module.

[0004] In the above-mentioned prior art, the emergency repair is evaluated and tested by collecting historical record information of emergency repair personnel repairing power failures of the same nature. However, the above-mentioned prior art lacks a solution for analyzing the improvement of staff capabilities and reallocating staff with improved capabilities. It is impossible to achieve continuous tracking and evaluation of staff capabilities based on actual work performance, which helps to identify high-performance staff and provide a basis for the optimal allocation of human resources. However, in the above-mentioned prior art, it is impossible to adjust the task allocation strategy in real time, ensure the flexibility and adaptability of task scheduling, and effectively improve the intelligence level of task scheduling.

[0005] To this end, the present invention provides an artificial intelligence task scheduling method based on deep learning. Summary of the invention

[0006] In order to make up for the deficiencies of the prior art, at least one technical problem raised in the background technology is solved.

[0007] The technical solution adopted by the present invention to solve the technical problem is: using a deep learning algorithm to analyze and process the task, performing deep learning on the task based on the performance data, processing to obtain the task evaluation value, and grading the task according to the task evaluation value; Based on the comparison between the task evaluation value and the task evaluation standard value, the task data is obtained, based on the task data, the task completion value and the advanced task completion value are obtained, the advanced performance value is processed to obtain the advanced performance value, the advanced performance value is compared with the advanced performance threshold, and the staff member is marked according to the comparison result; According to the comparison of senior performance values, the number of senior tasks and the number of senior personnel are compared to obtain the optimized scheduling value. The optimized scheduling value YH is combined with the senior performance value BX for analysis to obtain the optimized scheduling coefficient. According to the optimized scheduling coefficient XS, the senior tasks are reallocated to establish a reallocated task library. As a further technical solution of the present invention, the specific process of the deep learning algorithm is as follows: The deep learning task analysis algorithm based on feature extraction and multi-dimensional evaluation uses deep learning models (such as convolutional neural network CNN, recurrent neural network RNN, etc.) to extract features from task data. These features may include the type of task, the complexity of input data, the required computing resources, etc. The algorithm combines multi-task learning or transfer learning strategies to build a recommendation evaluation model. This model can comprehensively consider multiple dimensions of the task (such as urgency, difficulty, etc.), and finally output the evaluation results of the difficulty and priority of the task through training and optimization; As a further technical solution of the present invention, the comparison process of the task evaluation value is as follows: Obtain a task evaluation value according to the output of the recommended evaluation model, and compare the task evaluation value with the task evaluation standard value; If the task evaluation value is less than the task evaluation standard value, the task is judged as a low-level task; If the task evaluation value is greater than or equal to the task evaluation standard value, then the task is determined to be a high-level task; As a further technical solution of the present invention, the method for obtaining the advanced performance value is: Mark the task completion value as WC and the advanced task completion value as GJ; The task completion value and the advanced task completion value are analyzed and processed by the formula The advanced performance value BX is calculated, where s1 and s2 are both preset proportional coefficients; As a further technical solution of the present invention, the task completion value is obtained in the following manner: Set a preset period, and obtain the task completion amount and advanced task completion amount of a staff member within the preset period; The task completion amount of a staff member is calculated by ratio with the total amount of tasks in a preset period, and the task completion value is obtained by processing; As a further technical solution of the present invention, the method for obtaining the high-level task completion value is: The ratio of a staff member's high-level task completion amount to the task completion amount is calculated to obtain the high-level task completion value; As a further technical solution of the present invention, the advanced performance value is compared with the advanced performance threshold; If the advanced performance value is greater than the advanced performance threshold, the staff member is marked as advanced staff; If the high-level performance value is less than or equal to the high-level performance threshold, the staff member is marked as a low-level staff member; As a further technical solution of the present invention, the comparison process of the number of senior tasks and the number of senior personnel is as follows: Obtain the number of high-level tasks within the preset period through the task report of the task allocation department; Count the number of senior personnel and compare the number of senior personnel with the number of senior tasks; If the number of senior personnel is greater than or equal to the number of senior tasks, the senior tasks can be completed normally within the preset period; If the number of senior personnel is less than the number of senior tasks, the senior tasks cannot be completed on time within the preset period; As a further technical solution of the present invention, the method for obtaining the optimized scheduling value is: The difference between the number of senior personnel and the number of senior tasks within the preset period is calculated, and the absolute value is taken to obtain the optimized performance deviation. The ratio of the optimized performance deviation to the number of senior tasks is calculated to obtain the optimized scheduling value, which is marked as YH. As a further technical solution of the present invention, the method for obtaining the optimized scheduling coefficient is: Combine the optimized scheduling value YH with the advanced performance value BX for analysis, and use the formula The optimized scheduling coefficient XS is calculated, where a1 and a2 are both preset proportional coefficients.

[0008] The beneficial effects of the present invention are as follows: 1. Use deep learning algorithms to analyze and process tasks, perform deep learning on tasks based on performance data, obtain task evaluation values, and classify tasks according to the task evaluation values. The present invention uses deep learning algorithms to perform multi-dimensional evaluations on tasks, and can more accurately judge the difficulty and priority of tasks, thereby avoiding subjectivity and blindness in task allocation. According to the amount of tasks completed by staff members and the amount of high-level tasks completed, high-level performance values ​​are obtained through data analysis and processing, and then the staff members are marked as high-level or low-level. Based on the evaluation method of actual work performance, continuous tracking and evaluation of staff members' capabilities can be achieved, which helps to identify high-performance staff members and provide a basis for the optimal allocation of human resources; 2. According to the comparison of high-level performance values, the optimized scheduling value is obtained, and the optimized scheduling value YH is combined with the high-level performance value BX for analysis to obtain the optimized scheduling coefficient. The high-level tasks are reallocated according to the optimized scheduling coefficient XS, and a reallocated task library is established. In this way, reasonable task allocation and optimal allocation of human resources can be achieved, thereby improving overall work efficiency and quality. In addition, the embodiment of the present invention also provides a dynamic adjustment mechanism, that is, during the task execution process, the task allocation strategy is adjusted in real time according to the actual situation and the performance of the staff to ensure the flexibility and adaptability of task scheduling. The method provided by the embodiment of the present invention can effectively improve the intelligence level of task scheduling, and provide a new solution for the application of artificial intelligence in actual work. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The present invention will be further described below in conjunction with the accompanying drawings.

[0010] Figure 1 This is a flowchart of the steps of an artificial intelligence task scheduling method based on deep learning described in Example 1 of the present invention; Figure 2 This is a step flow chart of an artificial intelligence task scheduling method based on deep learning described in Example 2 of the present invention. DETAILED DESCRIPTION

[0011] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.

[0012] Example 1 like Figure 1 An artificial intelligence task scheduling method based on deep learning according to an embodiment of the present invention includes: Step 1: Use deep learning algorithms to analyze and process tasks, conduct deep learning on tasks based on performance data, obtain task evaluation values, and classify tasks according to the task evaluation values; The deep learning task analysis algorithm based on feature extraction and multi-dimensional evaluation uses deep learning models (such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs)) to extract features from task data. These features may include the type of task, the complexity of input data, the required computing resources, etc. The algorithm combines multi-task learning or transfer learning strategies to build a recommendation evaluation model that can comprehensively consider multiple dimensions of tasks (such as urgency, difficulty, etc.). Through training and optimization, it ultimately outputs a task evaluation value that evaluates the difficulty and priority of the task. Obtain a task evaluation value according to the output of the recommended evaluation model, and compare the task evaluation value with the task evaluation standard value; If the task evaluation value is less than the task evaluation standard value, the task is judged as a low-level task; If the task evaluation value is greater than or equal to the task evaluation standard value, then the task is determined to be a high-level task; It should be noted that the task evaluation value is an indicator to measure the importance of the task. In the recommended evaluation model, the larger the task evaluation value, the greater the priority and difficulty of the task, and it is marked as a high-level task. The smaller the task evaluation value, the smaller the priority and difficulty of the task, and it is marked as a low-level task. It needs to be explained that the task evaluation standard values ​​are set by the staff in this field through a large amount of historical data; Step 2: Based on the comparison between the task evaluation value and the task evaluation standard value, obtain the task data, obtain the task completion value and the advanced task completion value based on the task data, process them to obtain the advanced performance value, compare the advanced performance value with the advanced performance threshold, and mark the staff member according to the comparison result; It needs to be explained that the task data includes task completion and high-level task completion; Mark the task completion value as WC and the advanced task completion value as GJ; The task completion value and the advanced task completion value are analyzed and processed by the formula The advanced performance value BX is calculated, where s1 and s2 are both preset proportional coefficients; comparing the advanced performance value to an advanced performance threshold; If the advanced performance value is greater than the advanced performance threshold, the staff member is marked as advanced staff; If the high-level performance value is less than or equal to the high-level performance threshold, the staff member is marked as a low-level staff member; It should be noted that the advanced performance value reflects the degree to which the staff member completes advanced tasks. The larger the advanced performance value, the greater the proportion of advanced tasks in the tasks completed by the staff member, which can better highlight the ability of the staff member. The smaller the advanced performance value, the smaller the proportion of advanced tasks in the tasks completed by the staff member, which indicates that most of the tasks completed by the staff member are low-level tasks. The task completion value is obtained as follows: Set a preset period, and obtain the task completion amount and advanced task completion amount of a staff member within the preset period; It should be explained that the preset period is set by professionals in this field after precise calculation; The task completion amount of a staff member is calculated by ratio with the total amount of tasks in a preset period, and the task completion value is obtained by processing; It should be noted that the task completion value is a standard to measure the proportion of tasks completed by the staff member in the total tasks within the preset period. The larger the task completion value, the more tasks the staff member has completed within the preset period and the stronger his personal ability is. The method for obtaining the high-level task completion value is as follows: The ratio of a staff member's high-level task completion amount to the task completion amount is calculated to obtain the high-level task completion value; It should be noted that the high-level task completion value is the ratio of high-level tasks completed by the staff member to all the tasks completed by the staff member within the preset period. The larger the high-level task completion value is, the more high-level tasks the staff member has completed. The technical solution of the embodiment of the present invention is: use a deep learning algorithm to analyze and process tasks, perform deep learning on tasks based on performance data, obtain task evaluation values ​​through processing, and classify tasks according to the task evaluation values. The present invention uses a deep learning algorithm to perform multi-dimensional evaluation of tasks, which can more accurately judge the difficulty and priority of tasks, thereby avoiding subjectivity and blindness in task allocation. According to the staff's task completion volume and high-level task completion volume, a high-level performance value is obtained through data analysis and processing, and then the staff is marked as high-level or low-level. The evaluation method based on actual work performance can achieve continuous tracking and evaluation of staff capabilities, help identify high-performance staff, and provide a basis for the optimal allocation of human resources.

[0013] Example 2 like Figure 2 An artificial intelligence task scheduling method based on deep learning according to an embodiment of the present invention includes: Step 3: Based on the comparison of the senior performance values, the number of senior tasks and the number of senior personnel are compared to obtain the optimized scheduling value. The optimized scheduling value YH is combined with the senior performance value BX for analysis to obtain the optimized scheduling coefficient. The senior tasks are reallocated according to the optimized scheduling coefficient XS to establish a reallocated task library. Obtain the number of high-level tasks within the preset period through the task report of the task allocation department; Count the number of senior personnel and compare the number of senior personnel with the number of senior tasks; If the number of senior personnel is greater than or equal to the number of senior tasks, the senior tasks can be completed normally within the preset period; If the number of senior personnel is less than the number of senior tasks, the senior tasks cannot be completed on time within the preset period; In some embodiments, when the number of senior personnel is less than the number of senior tasks, the staff task allocation is optimized; Specifically, the difference between the number of senior personnel and the number of senior tasks within a preset period is calculated, and the absolute value is taken to obtain the optimized performance deviation, and the ratio of the optimized performance deviation to the number of senior tasks is calculated to obtain the optimized scheduling value, which is marked as YH; It should be noted that the optimized scheduling value YH reflects the degree of deviation between the number of senior personnel and the number of senior tasks. The larger the optimized scheduling value YH is, the greater the deviation between the number of senior personnel and the number of senior tasks is. The redistribution of the extra senior tasks can alleviate the pressure of task completion. Combine the optimized scheduling value YH with the advanced performance value BX for analysis, and use the formula The optimized scheduling coefficient XS is calculated, where a1 and a2 are both preset proportional coefficients; It should be noted that the optimization scheduling coefficient XS is calculated by combining the optimization scheduling value YH and the advanced performance value BX. The optimization scheduling coefficient XS reflects the optimization degree of advanced tasks, and the optimization scheduling value YH reflects the deviation between the number of advanced personnel and the number of advanced tasks. The larger the optimization scheduling value YH, the more it can reduce the pressure of completing tasks. The advanced performance value BX reflects the degree of completion of advanced tasks by staff members. The larger the advanced performance value, the greater the proportion of advanced tasks in the tasks completed by the staff member, which can better highlight the ability of the staff member. In some embodiments, high-level tasks are reallocated according to the optimized scheduling coefficient XS to establish a reallocated task library; In the redistributed task library, according to the size of the optimized scheduling coefficient XS, high-level tasks are preferentially assigned to workers with higher high-level performance values ​​to ensure efficient completion of tasks. At the same time, for those workers with lower high-level performance values, some low-level tasks can be appropriately assigned to improve their work enthusiasm and skills. Through the task management system, the progress and completion of tasks can be monitored in real time, and the extra senior tasks can be given priority to senior personnel who complete tasks faster. At the same time, tasks that exceed the capabilities of staff should be avoided to avoid task backlogs or reduced completion quality.

[0014] The technical solution of the embodiment of the present invention is: according to the comparison of high-level performance values, the optimized scheduling value is obtained, the optimized scheduling value YH is combined with the high-level performance value BX for analysis to obtain the optimized scheduling coefficient, the high-level tasks are reallocated according to the optimized scheduling coefficient XS, and a reallocated task library is established. In this way, reasonable allocation of tasks and optimal allocation of human resources can be achieved, thereby improving overall work efficiency and quality. In addition, the embodiment of the present invention also provides a dynamic adjustment mechanism, that is, during the task execution process, the task allocation strategy is adjusted in real time according to the actual situation and the performance of the staff to ensure the flexibility and adaptability of task scheduling. The method provided by the embodiment of the present invention can effectively improve the intelligence level of task scheduling, and provide a new solution for the application of artificial intelligence in actual work.

[0015] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. An artificial intelligence task scheduling method based on deep learning, characterized in that: include: Use deep learning algorithms to analyze and process tasks, conduct deep learning on tasks based on performance data, obtain task evaluation values, and classify tasks according to the task evaluation values; Based on the comparison between the task evaluation value and the task evaluation standard value, the task data is obtained, based on the task data, the task completion value and the advanced task completion value are obtained, the advanced performance value is processed to obtain the advanced performance value, the advanced performance value is compared with the advanced performance threshold, and the staff member is marked according to the comparison result; Based on the comparison of senior performance values, the number of senior tasks and the number of senior personnel are compared to obtain the optimized scheduling value. The optimized scheduling value YH is combined with the senior performance value BX for analysis to obtain the optimized scheduling coefficient. The senior tasks are reallocated according to the optimized scheduling coefficient, and a reallocated task library is established.

2. The artificial intelligence task scheduling method based on deep learning according to claim 1, characterized in that: The specific process of the deep learning algorithm is as follows: The deep learning task analysis algorithm based on feature extraction and multi-dimensional evaluation uses a deep learning model to extract features from task data. These features may include the type of task, the complexity of the input data, and the required computing resources. The algorithm combines multi-task learning or transfer learning strategies to build a recommendation evaluation model that can comprehensively consider multiple dimensions of the task. Through training and optimization, it ultimately outputs evaluation results on the difficulty and priority of the task.

3. The artificial intelligence task scheduling method based on deep learning according to claim 1, characterized in that: The comparison process of the task evaluation value is as follows: Obtain a task evaluation value according to the output of the recommended evaluation model, and compare the task evaluation value with the task evaluation standard value; If the task evaluation value is less than the task evaluation standard value, the task is judged as a low-level task; If the task evaluation value is greater than or equal to the task evaluation standard value, then the task is determined to be a high-level task.

4. The artificial intelligence task scheduling method based on deep learning according to claim 1, characterized in that: The advanced performance value is obtained as follows: Mark the task completion value as WC and the advanced task completion value as GJ; The task completion value and the advanced task completion value are analyzed and processed by the formula The advanced performance value BX is calculated, where s1 and s2 are both preset proportional coefficients.

5. The artificial intelligence task scheduling method based on deep learning according to claim 4 is characterized in that: The task completion value is obtained as follows: Set a preset period, and obtain the task completion amount and advanced task completion amount of a staff member within the preset period; The ratio of a staff member's task completion amount to the total amount of tasks within a preset period is calculated to obtain the task completion value.

6. The artificial intelligence task scheduling method based on deep learning according to claim 4 is characterized in that: The method for obtaining the high-level task completion value is as follows: The ratio of a staff member's high-level task completion amount to the task completion amount is calculated and processed to obtain the high-level task completion value.

7. The artificial intelligence task scheduling method based on deep learning according to claim 1, characterized in that: comparing the advanced performance value to an advanced performance threshold; If the advanced performance value is greater than the advanced performance threshold, the staff member is marked as advanced staff; If the high-level performance value is less than or equal to the high-level performance threshold, the staff member is marked as a low-level staff member.

8. The artificial intelligence task scheduling method based on deep learning according to claim 1, characterized in that: The comparison process of the number of senior tasks and the number of senior personnel is as follows: Obtain the number of high-level tasks within the preset period through the task report of the task allocation department; Count the number of senior personnel and compare the number of senior personnel with the number of senior tasks; If the number of senior personnel is greater than or equal to the number of senior tasks, the senior tasks can be completed normally within the preset period; If the number of senior personnel is less than the number of senior tasks, the senior tasks cannot be completed on time within the preset period.

9. The artificial intelligence task scheduling method based on deep learning according to claim 1, characterized in that: The method for obtaining the optimized scheduling value is: The difference between the number of senior personnel and the number of senior tasks within the preset period is calculated, and the absolute value is taken to obtain the optimized performance deviation. The ratio of the optimized performance deviation to the number of senior tasks is calculated to obtain the optimized scheduling value, which is marked as YH.

10. The artificial intelligence task scheduling method based on deep learning according to claim 1, characterized in that: The optimization scheduling coefficient is obtained as follows: Combine the optimized scheduling value YH with the advanced performance value BX for analysis, and use the formula The optimized scheduling coefficient XS is calculated, where a1 and a2 are both preset proportional coefficients.

Images