基于LMDP的铁路钢轨网格化维修规划编制模型

LMDP-based Optimal Repair Planning Model for Railway Rail Grids

铁路钢轨设备维修规划的优化编制,能够辅助合理分配维修资源与控制维修成本。构建了一种新的铁路钢轨维修规划编制模型。按照某一长度分割铁路线路,得到多个连续等长的小钢轨区段,这些小钢轨区段称为钢轨网格。以钢轨网格为研究对象,采用Markov随机过程理论,个性化地描述了钢轨网格状态劣化规律,利用隐形马尔科夫决策过程方法,建立了基于状态的维修策略模型,以一个规划周期内的期望总成本最小为优化目标,确定规划周期内的最优检查策略、维修策略,并采用陇海线的实际数据,对提出模型的有效性进行了验证,计算结果表明提出的模型优于既有的基于阈值的管理方法。

Optimal repair planning for railway rail contributes to proper allocation of maintenance resources and effective control of maintenance costs. This paper proposed a novel optimal repair planning model for railway rails. Based on the same length, a railway line was separated into a plurality of adjacent small segments. Each rail segment was termed a “rail grid”. Rail grids were taken as the research subjects. Employing the theory of Markov stochastic processes, the degradation law of each rail grid was customized. Latent Markov decision process was employed to establish a condition-based repair policy model, and find out joint optimal inspection and repair policies over a finite planning horizon to minimize the total expected cost in a planning cycle. The actual data for the Lanzhou-Lianyungang Railway was used to verify the proposed model. The calculation results show that the proposed model is superior to the conventional threshold-based methods.