Manufacture cloud service execution time prediction method and manufacture cloud service execution time prediction device

Abstract

A method and device for predicting the execution time of manufacturing cloud services. When predicting the service execution time of each candidate cloud service, first determine the characteristic mode of the service execution time of the candidate cloud service, and then according to the set characteristic mode and prediction For the corresponding relationship of the model, use the determined predictive model corresponding to the feature pattern to calculate the predicted value of the service execution time of the candidate cloud service; wherein, the predictive model adopts an exponential smoothing algorithm with adaptive smoothing coefficients for prediction. The present invention improves the traditional exponential smoothing prediction method; proposes an identification method for the characteristic pattern of manufacturing cloud service execution time, and predicts the execution time of manufacturing cloud service on this basis, which enhances the adaptability of the prediction method and improves the prediction Accuracy can provide important data support for service combination, resource optimization configuration and management in cloud manufacturing environment.

Description

technical field [0001] The present invention relates to the field of information technology, in particular to a method and device for predicting service execution time in a cloud manufacturing environment. Background technique [0002] The manufacturing industry occupies an important position in the national economy, and the traditional manufacturing industry is facing great challenges and new opportunities in the information age. The networked manufacturing model has become a new choice for the manufacturing industry, but there are many problems in the networked manufacturing models such as manufacturing grid and agile manufacturing in terms of service model, manufacturing resource sharing and allocation, physical terminal equipment access, and information security. In this context, cloud manufacturing came into being. Cloud manufacturing connects manufacturing resources to the network through embedded, Internet of Things and other technologies, organizes online manufactur...

AI Technical Summary

Problems solved by technology

However, the above research work is mainly concentrated on the grid computing environment, and the cloud manufacturing platform has its own unique characteristics such as physical device resource access, resource heterogeneity, and organizational dynamic changes, which cannot be applied to the above methods.

[0005] In addition, the existing service execution time prediction technology rarely analyzes the predicted objects. There are a large number of cloud services that can complete the tasks requested by users in the cloud manufacturing environment. If the execution time prediction is performed for all cloud services, a lot of time will be wasted and cost

Images