Task distribution method and device, electronic equipment and computer readable storage medium

Abstract

The embodiment of the invention discloses a task allocation method and device, electronic equipment and a computer readable storage medium. the task method comprises the steps: obtaining task-relatedtraining data, and carrying out the training according to the task-related training data, and obtaining a task execution cost prediction model; Obtaining related data of a to-be-distributed task, andinputting the related data into the task execution cost prediction model to obtain a predicted execution cost of the to-be-distributed task; And according to the predicted execution cost of the to-be-allocated task, allocating the to-be-allocated task. According to the technical scheme, the input of manpower can be greatly reduced, the working efficiency can be improved, the optimization of task distribution can be effectively ensured, the distribution risk can be reduced to the maximum extent, and meanwhile, the machine learning method is applicable to many scenes due to the fact that the training data is complete and the self-adaptive capacity is high.

Description

technical field [0001] The present disclosure relates to the technical field of data processing, and in particular to a task allocation method, device, electronic equipment, and computer-readable storage medium. Background technique [0002] With the development of Internet technology, more and more merchants or service providers provide services to users through Internet platforms. In order to improve service quality and user experience, service providers usually optimize the allocation of service tasks such as orders. In the prior art, there is an optimal processing method for task allocation, which is to first use some preset rules to score the cost between delivery personnel and other order delivery resources and orders, and combine the obtained scores between delivery resources and orders into two components: Dimensional matrix, and then use the KM algorithm to perform maximum weight matching based on the score two-dimensional matrix, and then optimize the task allocat...

AI Technical Summary

Problems solved by technology

However, in the prior art, the two-dimensional score matrix is ​​usually generated by manual rules and manual experience, which requires a lot of manpower to practice in the market, and then modify and adjust the current scoring rules after getting feedback. and then put into the market

Such non-stop iteration not only leads to waste of manpower and low efficiency, but also requires certain risks for each experiment. In addition, the logic of the above manual scoring rules is relatively simple, and it is difficult to apply to all scenarios

Images