������ɢ����µĹ��������мƻ��Ż��о�

提出飓风等自然灾害条件下运用公交车进行居民紧急疏散的优化模型. 最优公交车疏散运行计划问题可转化为不确定性需求的选址—路径优化模型，目标函数是使总疏散时间最小. 选址—路径优化模型用于确定最有效的公交车集结点服务区域和将人员从受灾区域转移到指定避难所或安全地区的最优线路，并设计遗传算法、神经网络算法和爬山算法结合的混合启发式算法. 通过美国密西西比州格尔夫波特市的实际数据对所提出的模型进行验证. 实验结果表明，混合遗传算法在求解效果和效率上都优于传统的遗传算法.

Optimum Transit Operations during the Emergency Evacuations

This paper presents an optimization modeling technique to develop an evacuation plan for transit-dependent residents during in the event of natural disaster such as the emergency hurricane situation. The transit evacuation operation problem is formulated as a class of location-routing problem (LRP) with uncertain demands. The objective function is set to minimize the total evacuation time. The LRP problem identifies the optimal serving areas and transit vehicle routings to move evacuees from the flooding affected zone to designated shelters or safe destinations. The computational experience is presented on the application of hybrid genetic algorithms, artificial neural network, and hill climbing heuristic algorithms. Numerical experiments are conducted using the real survey data from Gulfport, MS, to illustrate the proposed modeling technique. Experimental results show that the hybrid GA performs well both in quality and efficiency than traditional GA.