Intelligent Algorithms On The Stochastic Loader Problem

Some algorithms, such as simulated annealing algorithm(SA), genetic algorithm(GA), artificial neural networks, ant colony optimization(ACO) and particle swarm optimization(PSO) is the new intelligent algorithms to solve problems, especially for the NP-hard problems of optimization. With the successful application in practical optimization, intelligent algorithms have become a powerful tool to solve the complicated optimization problem. Now it is the frontiers of research in algorithms and artificial intelligence fields.In this paper, we present a new genetic algorithm with improved evolutionary computing and a new particle swarm optimization, then use them to solve stochastic loader problem.The innovative viewpoints of this dissertation are as follows:1. On the basis of genetic algorithm, this paper presents a new hybrid genetic algorithm for the stochastic problem, this algorithm combines new crossover and mutation operators so that it has faster convergence speed in comparison with the basic GA.2. The algorithm is based on the competition among individuals, so we give an improved hybrid genetic algorithm which involves learning process on the basis of the new algorithm given above, and the improved algorithm introduce the cooperate and learning system among individuals. Numerical examples illustrated the improved genetic algorithm has the acceptable accuracy and the faster convergence in solving the stochastic loader problem..3. The basic PSO is a kind of algorithm in continuous space, since the stochastic problem belongs to integer programming (IP), we adjust and modify the PSO algorithm process, which obtains the higher solutionprecision and convergence efficiency.