Research And Application Of Resource Scheduling In Network Operation And Maintenance Based On Genetic Algorithm

In recent years, with the development of communication market competition is becoming increasingly fierce, especially in the 4G era, reduce network maintenance costs become the important means of enterprises to enhance the communication of weapons. According to the Ministry of industry in 2014 statistical data show that, the three operators of the total number of 4G base station has reached 70 million, network operators become more and more complex, operation and maintenance of the workload is more and more big, network operation and maintenance cost is also getting higher and higher. Therefore, reducing the cost of network operation and maintenance has become the focus of attention of the communications business. The field operation of the network operation and maintenance scheduling cost is an important aspect of network operation and maintenance costs, according to the requirements of operation order appropriate allocation of resources to the task, to provide better network operation and maintenance service quality, this is a problem needs to be solved in the field task scheduling. Because there are a lot of resources, involving network operation field, dynamic scheduling scheme for real-time dynamic adjustment, the traditional manual scheduling is facing a huge challenge. Therefore, how to design a good site scheduling algorithm, the scheduling to reduce costs as much as possible, improve the network operation quality of service, is an important and realistic problem.According to the characteristics of network operation and maintenance on-site job scheduling problem, this paper proposes combining the genetic algorithm and simulated annealing algorithm to solve the problem of resource scheduling algorithm, namely simulated annealing genetic algorithm SAGA(simulatedannealing genetic algorithm). First, a detailed description of the definition of network operation and maintenance in the field scheduling problems, the job scheduling problem on site analysis and on-site job scheduling model, respectively, using genetic algorithm and simulated annealing genetic algorithm to solve the job scheduling model.Genetic algorithm is global optimization search algorithm. However, it has easily trapped into local optimal solution and slow convergence speed etc., and simulated annealing algorithm has characteristics of escaping from the local optimal solution. In this paper, the combination of the two algorithms, give full play to their advantages. Using SAGA to solve the problem of resource scheduling in the field work is divided into two stages:the genetic operation stage and the simulated annealing stage. First, describes the use of genetic algorithm to solve the scheduling problem of field network operation when encoding, selection, crossover, mutation and fitness function design, according to the characteristics of field network operation and maintenance scheduling, genetic operation phase encoding using the double encoding, the field operation engineers to complete the task of encoding and field operations need to use the resources of the engineer encoding; secondly, of genetic operation stage, obtained the solution by introducing the simulated annealing algorithm optimization, populations in genetic operation based, using simulated annealing operation, at a specific temperature, through to each individual of the population of the metropolis pick by the process and the formation of new species, continue to iterate and, ultimately, to find the optimal solution.The scene scheduling solution model is applied to Guangzhou on behalf of a dimension of the enterprise on-site job scheduling and to solve on-site job scheduling model of saga in the enterprise development of intelligent maintenance service management platform, effectiveness of using an example to validate that the saga in solving the network operation and maintenance in the field scheduling problem.