Abstract: As a particular NP-C problem, the train operation adjustment is hard to obtain optimal excellent solution because of numerous constraints, the huge search space and the narrow area of the feasible solution. According to the characteristics of high speed train operation adjustment, a Discrete Firefly Algorithm（DFA） is put forward to solve the practical problem, which is based on the Firefly Algorithm（FA） that is representative in the intelligent algorithm. The disturbance mechanism based on variable neighborhood search algorithm is used to increase the diversity of the fireflies group and avoid algorithm trapped in local optimal solution. The algorithm will be used in high-speed train operation adjustment problem, and the calculation results of adjustment scheme based on discrete firefly algorithm is better than the general heuristic adjustment algorithm through the analysis of example.

Key words: train operation adjustment, discrete firefly algorithm, variable neighborhood search

摘要： 列车运行调整是一类特殊的NP完全问题，由于约束众多，搜索空间庞大，可行解范围狭小，因此难以获得最优解。针对高速列车运行调整问题的特点，以智能算法中有代表性发展优势的萤火虫算法（FA）为基础，根据实际问题提出一种离散的萤火虫算法（DFA）进行求解。为了增加萤火虫群的多样性，避免算法陷入局部最优解，采用了基于变邻域搜索算法的扰动机制。将该算法用于高速列车运行调整问题，经过算例对比分析，基于离散萤火虫算法调整方案的计算结果优于普通启发式算法调整结果。

关键词: 列车运行调整, 离散萤火虫算法, 变邻域搜索