Multi-agent Scheduling Optimization Algorithm

Scheduling is a form of decision-making in military, aerospace, energy, industrial manufacturing and agricultural production ,and it plays an important role in many areas. At present the main problem of scheduling is how to generate an actual available scheduling system in reality. There are more segmentation research directions, and most of them are focused on the specific scheduling environment and the scheduling objective within the framework of finding the optimal scheduling solution. In particular, with the development of various types of new algorithms and scheduling policy and rules to meet the prerequisite, improving the scheduling algorithm to obtain the optimal scheduling scheme has become an important method to combine oretical research and application.In this paper,the multi-agent technology to establish the form models is studied,which can be used to perform a specific task scheduling objective in the future large surface ships. In the wide constraints, the use of genetic algorithm is a feasible schedule.Centralized control and distributed control are the main types of task scheduling system control frame structure . Since the characteristics of timing series in the scheduling syetem of discrete tasks scheduling systems, the mathematical model of the system and the necessary constraints are worked out by combining the serial and the parallel task scheduling modes. Based on intelligent agent technology , a form models of multi-agent task scheduling system is established.The basic task scheduling overview of genetic algorithms is introduced. Based on sub-structure of chromosome coding ideas, task scheduling objective function is proposed to design three kinds of genetic operators and auxiliary correction operators by using the dynamic characteristics of generating feasible task scheduling algorithm,and the simulation results show the effectiveness of the proposed generic algorithm.For unfeasible schedule may be generated in coding segments of standard genetic algorithm,a new method is brought out,which is combined with genetic algorithm and local heuristic search algorithm. By the method of piecewise coding which obeys the priority rules, the times that unfeasible schedule can be reduced, and the reliability of generating scheduling scheme and the efficiency in the implementation can be improved.The problems that how the number of the nodes of the task will be changed and how the execution order of the tasks will be specified in scheduling task can also be solved by the algorithm perfectly.