改进柔性资源约束项目调度模型与粒子群算法

Modified Model of Flexible Resource-constrained Multi-mode Project Scheduling Problem and Particle Swarm Optimization

在资源受限项目调度问题中,将可再生资源进一步拓展为具有能力差异的柔性资源,建立考虑能力差异的柔性资源受限的多模式项目调度问题模型,该模型是对传统资源约束项目调度问题(RCPSP)更接近实际的拓展。提出了基于粒子群算法的求解算法,粒子群算法求解该模型的思路为,利用蒙特卡洛方法根据资源-能力矩阵与活动模式-能力矩阵得到活动模式-资源矩阵,将考虑能力差异的柔性资源受限的多模式项目调度问题转换为常规的多模式项目调度问题,然后利用基于任务序列与模式表示的粒子群算法对该多模式项目调度问题进行求解。用数值实例说明了模型的合理性与算法的有效性。

Along with the resource in multi-mode resource-constrained project scheduling problem being extended to flexible resource with capability difference,a mix-integer programming model is established for modified flexible resource-constrained multi-mode project scheduling problem （MFRCMPSP） considering capability difference; the modified model is a practical extension of traditional resource-constrained project scheduling problem （RCPSP）. The concept of particle swarm optimization for resolving the model is that getting activity-mode-resource matrix basing on resource- capability matrix and activity-mode-capability matrix with Monte Carlo methods, and transforming the model of MFRCMPSP considering capability difference into traditional multi-mode resource- constrained project scheduling problem （MRCPSP）, and using particle swarm optimization （PSO） in which particle representation is based on permutation and mode to resolve the MRCPSP. Results of example showed that model is reasonable,and this considering capability difference. algorithm could effectively solve FRCMPSP