An Immune Algorithm Based Robust Scheduling Methods

Inspired from the robust control principle, a robust scheduling method is proposed to solve uncertain scheduling problems. The uncertain scheduling problem is modeled by a set of workflow simulation models, and then a scheduling scheme (solution) is evaluated by the results of workflow simulations that are executed by using the workflow models in the set. A variable neighborhood immune algorithm (VNIA) is used to obtain an optimal robust scheduling scheme that has good performances for each model in the model set. The detailed steps of optimizing robust scheduling scheme by the VNIA are given. The antibody coding and decoding schemes are also designed to deal with resource conflicts during workflow simulation processes. Experimental results show that the proposed method can generate robust scheduling schemes that are insensitive for uncertain disturbances of scheduling environments.