Two-level multi-task scheduling in a cloud manufacturing environment

Cloud manufacturing is a new paradigm of global manufacturing networks. In a cloud manufacturing environment, the problem of scheduling multiple heterogeneous tasks is very challenging. Tasks submitted to the cloud manufacturing platform possess different functional requirements and various complexities. They are decomposed into fine-grained subtasks with their precedence relationship to be processed by the aggregated distributed resources. From the perspective of each individual task, it is desirable to be completed in the shortest time possible without due date constraint. On the other hand, from the perspective of the cloud platform, all tasks being scheduled should also be completed as quickly as possible. How to schedule all the subtasks decomposed from multiple heterogeneous tasks to maximize the benefits not only to individual tasks themselves but also to the platform as a whole is an intractable but important issue in cloud manufacturing. In this paper, a two-level multi-task scheduling model is proposed. Two scheduling strategies based on the two-level scheduling model are presented and evaluated. The test results obtained from computational experiments suggest that the proposed two-level scheduling model is more effective than the traditional one-level model. The two-level two-optimization scheme in particular can find a better schedule beneficial not only to individual task but also to the cloud platform as a whole.