军事物流选址分配模型及遗传算法优化

在每个物流中心的容量有限的情况下,为了解决军事物流中心选址和服务需求分配问题,建立了一种基于二维地理坐标平面的军事物流选址分配模型。针对这种非凸和非光滑的混合整数非线性规划模型,给出了一种改进遗传算法的全局最优化求解方法,其中由于适应度函数与各物流中心对应的需求分配情况密切相关,我们用拉格朗日松弛法解决了对于特定位置的物流中心服务需求分配的子问题,最终得到了模型的全局最优解或者近似全局最优解,最后的实例求解表明了该算法的有效性和稳定性。

Optimization of the Model of Military Logistics Center Location and Allocation of Service Using Genetic Algorithm

On condition that each logistics center facility has a service capacity,a model of military Logistics Center Location and Allocation is presented based on planar geography coordinates to solve the locations of centers and the allocations of service demands.To deal with the non-convex and non-smooth Mixed Integer Nonlinear Programming model,an improved enetic algorithm of global optimization solution is given,and the sub-problem of allocations of service demands is solved by Lagrangian Relaxation Method,finally we can get a global optimal solution or a nearly global optimal solution.Its validity and stability are testified by a practical example.