基于随机规划的机场选址模型研究

The Optimization of Airport Site Selection Based on Stochastic Programming

随着中国经济的快速发展,中国民航业正在迅速崛起,越来越多的机场正在规划建设当中。机场选址作为机场建设的前期工作,是机场规划建设的一个重要组成部分。现有的机场选址理论大都基于单个新建机场自身出发,很少考虑同一个城市群当中其他机场的竞争与合作。将同一个城市群内的所有机场作为一个整体进行考虑,深入研究新建机场的选址问题。首先,从中国交通发展规划的要求出发,指出民航主要承担不同城市群间城市的旅客运输服务。随后建立了以旅客综合出行成本最低为目标函数的机场选址数学模型,该模型为非线性整数随机规划模型。设计了两阶段算法对模型进行求解。即采用枚举法将非线性规划问题转换成多个线性整数随机规划子问题。然后采用蒙特卡罗方法将随机规划子问题转化为线性整数规划问题,再采用分支定界算法进行求解得到解集后,采用统计分析和随机抽样的方法确定最优解。最后采用算例证明了该模型和求解算法的有效性和实用性。

With the rapid development of China＇s economy,civil aviation in China is rapidly rising,more and more airports need to build up. The airport site selection as an early work of the airport construction,is an important part of the airport planning and construction. The existing airport site selection is mainly based on single newbuilt airport itself,less to consider the competition and cooperation with the other airports in the same urban agglomeration. all the airports in the same region as a whole was consider,fully studying the competition and coordination among different airports. Fistly from the analysis about the requirements of China＇s comprehensive transportation development,civil aviation mainly bear passenger transportation service to another urban aggregation or abroad. And then airport layout optimization mathematical model is built up,which is a nonlinear integer stochastic programming model with minimum comprehensive trip cost of passenger as objective function. In order to solve the model,a two-stage algorithm is designed. enumeration method is be introduced into transforming original problem into multiple linear integer stochastic programming sub-problems. The subproblem is simplified to linear integer programming problem by monte carlo method,so that branch-and-bound can be used. At last,a numeric example is made to demonstrate the method proposed is effective and feasible.