Wang Dongyan, Sun Ying, Yu Liang, Shen Kun, Li Junbo, Wu Xia
Institute of Computing Technology, China Academy of Railway Sciences Group Co., Ltd., Beijing, China.
School of Civil Engineering, Central South University, Changsha, China.
PLoS One. 2025 Jul 30;20(7):e0327852. doi: 10.1371/journal.pone.0327852. eCollection 2025.
As cities grow, intercity railways are becoming increasingly popular for short trips between neighboring areas. These railways cater well to commuters and travelers, making reliable and cost-effective maintenance crucial. Timely access to spare parts is essential for ensuring the smooth operation of intercity railways. Traditionally, intercity railways lack failure probability data for spare parts, which hampers the support for spare parts ordering decisions, resulting in spare parts management primarily relying on manual experience. This approach often leads to problems like excessive inventory levels and high management costs. To enhance the reliability of intercity railway operations and reduce spare parts management costs, this paper employs the Zebra Optimization Algorithm-Least Squares Support Vector Machine (ZOA-LSSVM) to analyze the reliability of the important Weibull distribution spare parts of the intercity railway and fit the parameters of the reliability function for spare parts. Based on the failure rate, an inventory control model for intercity railway spare parts is established, aiming to minimize total costs while considering constraints such as order point, order quantity, and equipment availability. A genetic algorithm is designed to solve this model. To verify the effectiveness of the model, we select the contact network insulators of Chinese J Intercity Railway as the case study subject. By comparing the fitting performance of several methods, including ZOA-LSSVM, Genetic Algorithm (GA)-LSSVM, LSSVM, and Least Squares Regression (LSR), the effectiveness of ZOA-LSSVM is validated. The experimental results indicate that ZOA-LSSVM can provide better prediction accuracy. Based on this fitting method, spare parts inventory management is conducted. By comparing it with the traditional manual experience method, it is found that the approach proposed in this paper not only ensures the stable operation of intercity railways but also significantly reduces costs by approximately 13.6%. This result fully demonstrates the superiority of the optimization model established in this paper in practical applications and provides new ideas and methods for the management of spare parts for other intercity railways.
随着城市的发展，城际铁路在相邻地区之间的短途出行中越来越受欢迎。这些铁路很好地满足了通勤者和旅行者的需求，因此可靠且经济高效的维护至关重要。及时获取备件对于确保城际铁路的平稳运行至关重要。传统上，城际铁路缺乏备件的故障概率数据，这阻碍了对备件订购决策的支持，导致备件管理主要依赖人工经验。这种方法常常导致库存水平过高和管理成本高昂等问题。为了提高城际铁路运营的可靠性并降低备件管理成本，本文采用斑马优化算法-最小二乘支持向量机（ZOA-LSSVM）来分析城际铁路重要威布尔分布备件的可靠性，并拟合备件可靠性函数的参数。基于故障率，建立了城际铁路备件的库存控制模型，旨在在考虑订单点、订单数量和设备可用性等约束条件的同时，使总成本最小化。设计了一种遗传算法来求解该模型。为了验证模型的有效性，我们选择中国J城际铁路的接触网绝缘子作为案例研究对象。通过比较包括ZOA-LSSVM、遗传算法（GA）-LSSVM、LSSVM和最小二乘回归（LSR）在内的几种方法的拟合性能，验证了ZOA-LSSVM的有效性。实验结果表明，ZOA-LSSVM能够提供更好的预测精度。基于这种拟合方法，进行了备件库存管理。将其与传统的人工经验方法进行比较，发现本文提出的方法不仅确保了城际铁路的稳定运行，还显著降低了成本，降幅约为13.6%。这一结果充分证明了本文建立的优化模型在实际应用中的优越性，并为其他城际铁路的备件管理提供了新的思路和方法。
