论文部分内容阅读
本文提出了一种解决设备更新换代优化(ERO)问题的随机动态规划(SDP)模型,用以明确地解释在车辆利用中的不确定性,并采用Bellman算法解决ERO SDP问题.针对SDP状态空间的增长,提出了特殊简化算法,以解决动态规划方法中固有的“维数灾”问题,确保所需的内存和计算时间不会随着时间范围的增加而成倍增长.并对SDP软件的实现技术、功能和图形用户界面(GUI)进行了讨论,开发了基于SDP的ERO软件,并使用美国得克萨斯交通局(TxDOT)现有车辆数据进行验证.对统计结果、软件计算时间和求解效果进行综合分析,结果显示,使用该ERO软件,估计大量成本可以节省.
In this paper, we propose a stochastic dynamic programming (SDP) model to solve the equipment replacement optimization (ERO) problem, which can explain the uncertainty in vehicle utilization and solve the ERO SDP problem using Bellman algorithm.According to the SDP state space , We propose a special simplification algorithm to solve the problem of “dimensionality” inherent in the dynamic programming method to ensure that the required memory and computing time do not increase exponentially with the increase of the time range. Software implementation techniques, functionalities, and graphical user interface (GUI) are discussed, and SDP-based ERO software is developed and validated using existing vehicle data from the U.S. Texas Department of Transportation (TxDOT). The statistical results, software calculation time and solution The results of a comprehensive analysis, the results show that the use of the ERO software, estimated cost savings can be substantial.