The Research On Aircrew Scheduling For Multi-Objective Optimization

Aircrew scheduling is an important part of airlines production tasks. It not only must follows the provisions of the civil aviation administration, but also needs comprehensive consideration working hours and job intensity of the steward and other factors. It has features of multivariable, strong coupling, multi-objective and so on. Aircrew assignment is one process of comprehensive optimization for multiple targets. However, the research of aircrew scheduling mainly concentrated in single-objective optimization. Usually several of objectives belong to different dimension, and even conflicting. Consequently, converting the multi-objective scheduling problem into single-objective optimization problem can not solve the practical question.In this paper, a multi-objective model of aircrew scheduling which adapt to our nation is presented; multiple targets will be optimized at the same time. After the characteristics of aircrew scheduling problem and the advantages and disadvantages of various algorithms are analyzed, two method, chaos immune algorithm and immune particle swarm algorithm, are improved and selected to optimize the scheduling process. Chaos technology is used to produce non-repeating sequence and initialize the group in the first iteration of immunity arithmetic, as it is difficult to find a better one from a large number of randomly solutions in a short time. Particle swarm algorithm has higher search efficiency, but easy to fall into local optimum. So, the immune memory and the selection mechanism which based on concentration are introduced into particle swarm algorithm.In the end of this paper, the multi-objective model for crew scheduling, chaos immune algorithm and immune particle swarm algorithm applied to the aircrew scheduling system, the required aircrew scheduling results verify the legitimacy and effectiveness of the algorithm.