An Intelligent Genetic Scheme for Multi-Objective Collaboration Services Scheduling

The optimization of collaborative service scheduling is the main bottleneck restricting the efficiency and cost of collaborative service execution. It is helpful to reduce the cost and improve the efficiency to deal with the scheduling problem correctly and effectively. The traditional genetic algorithm can solve the multi-objective problem more comprehensively than the optimization algorithm, such as stochastic greedy algorithm. But in the actual situation, the traditional algorithm is still one-sided. The intelligent genetic scheme (IGS) proposed in this paper enhances the expansibility and diversity of the algorithm on the basis of traditional genetic algorithm. In the process of initial population selection, the initial population generation strategy is changed, a part of the population is randomly generated and the selection process is iteratively optimized, which is a selection method based on population asymmetric exchange to realize selection. Mutation factors enhance the diversity of the population in the adaptive selection based on individual innate quality. The proposed IGS can not only maintain individual diversity, increase the probability of excellent individuals, accelerate the convergence rate, but also will not lead to the ultimate result of the local optimal solution. It has certain advantages in solving the optimization problem, and provides a new idea and method for solving the collaborative service optimization scheduling problem, which can save manpower and significantly reduce costs on the premise of ensuring the quality. The experimental results show that Intelligent Genetic algorithm (IGS) not only has better scalability and diversity, but also can increase the probability of excellent individuals and accelerate the convergence speed.

Automated summary

The optimization of collaborative service scheduling is crucial for improving efficiency and reducing costs. This paper proposes an Intelligent Genetic algorithm (IGS) which enhances the scalability and diversity of the algorithm. By changing the initial population generation strategy and implementing an adaptive selection based on mutation factors, the IGS is able to maintain individual diversity, accelerate convergence speed, and avoid local optimal solutions. The experimental results demonstrate that IGS has better scalability and diversity, while increasing the probability of excellent individuals and accelerating convergence speed.