基于资源流网络优化的鲁棒性项目调度

针对项目执行过程中由于活动拖期导致基准调度计划不断变更的问题，从资源分配的角度构建基于资源流网络优化的鲁棒性调度计划。首先设计拖期惩罚成本指标来衡量调度计划的鲁棒性，并构建以拖期惩罚成本最小化为目标的资源流网络优化动态模型。针对该模型设计MTPC资源流网络优化算法，该算法以活动为基准，采用拖期惩罚成本最小的资源分配方案实现资源在活动节点之间的有效流动，提升调度计划的鲁棒性。最后，为验证MTPC优化算法的有效性和可行性，通过采用蒙特卡罗模拟仿真实验将MTPC优化算法与RRAS，Min-EA和MABO等3种资源分配算法进行对比分析。实验结果表明：MTPC算法在调度计划的鲁棒性，资源分配方案的稳定性以及算法的时间效率上都优于其他3种算法。MTPC算法不仅能快速有效地完成资源配置，还能通过降低活动的拖期风险提升调度计划的鲁棒性，这可以帮助项目管理者构建抗干扰能力较强的基准调度计划。

Robust Project Scheduling Based on Optimization of Resource Flow Network

Aimed at the fact that project schedules have to be re-scheduled owing to tardiness in activities,robust baseline schedules are generated based on optimization of the resource flow network from resource allocation.First,a tardiness penalty cost(TPC)indicator is proposed to measure the schedule robustness,based on which,a dynamic model of resource flow network optimization is developed to minimize the sum of TPC.Next,a minimizing tardiness penalty cost(MTPC)algorithm is developed for solving the model.This algorithm is activity-based,in which resources are efficiently transferred from one activity to the other with a minimal TPC in order to increase the schedule robustness.Finally,Monte Carlo computational experiments are conducted to verify the feasibility and effectiveness of the MTEC algorithm compared with other three algorithms(RRAS,Min-EA and MABO).The computational results indicate that the MTEC algorithm outperforms these three algorithms in schedule robustness,stability of resource allocation,and temporal efficiency of algorithms.Therefore,the MTPC can not only efficiently allocate resources,but also improve the schedule robustness by reducing the tardiness in activities,which helps project managers to develop more stable baseline schedules against disruptions.