摘要
以最大化现金流净现值为优化目标的多模式资源约束调度问题MMRCPSP(Multi-mode Resource-Constrained Project Scheduling Problem)是一类带有复杂非线性特征的NP-hard问题,传统粒子群算法在解决该类离散问题上具有一定局限性。从粒子群算法的优化原理出发,结合遗传算法,在粒子群算法中引入交叉和变异操作,得出一种应用于MMRCPSP现金流优化的快速、易实现的混合粒子群算法,拓宽了粒子群优化算法在离散优化领域的应用。仿真实验结果验证了算法的有效性和高效性。
Multi-mode resource constrained project scheduling （MMRCPSP） is an NP-hard problem with complex non-linear feature, which has the optimisation object of maximising the cash flow net present value. Traditional particle swarm optimisation has certain limitation in solving such discrete problems. Proceeding from the optimisation principle of PSO and combining the genetic algorithm, in this article we introduce crossover and mutation operation to PSO and derive a hybrid PSO to be applied to MMRCPSP cash flow optimisation, it is fast and easy to use, and widens the application of PSO. in discrete optimisation field. Simulation experiment results demonstrate the validity and efficiency of this algorithm.
出处
《计算机应用与软件》
CSCD
2010年第3期275-278,共4页
Computer Applications and Software
关键词
粒子群算法
现金流优化
净现值
交叉
变异
Particle swarm optimisation （PSO） Cash flow optimisation Net present value Crossover Mutation
