Complementarity Active-Set Algorithm for Mathematical Programming Problems with Equilibrium Constraints

In this paper, an algorithm for solving a mathematical programming problem with complementarity (or equilibrium) constraints (MPEC) is introduced, which uses the active-set methodology while maintaining the complementarity restrictions throughout the procedure. Finite convergence of the algorithm to a strongly stationary point of the MPEC is established under reasonable hypotheses. The algorithm can be easily implemented by adopting any active-set code for nonlinear programming. Computational experience is included to highlight the efficacy of the proposed method in practice.     

车辆路径问题的混合优化算法

讨论了一类车辆路径调度问题(VRP)及其数学模型，并且分析了以遗传算法求解该类问题时的染色体表示和有关遗传操作，然后结合2-opt局部优化算法提出了GA with2-opt算法来求解VRP问题，试验结果说明了该算法的有效性和可行性。     

遗传算法的物理解释

提出了一种理想化的模拟仿生搜索算法——扰动算法 ,以此方法为基础 ,分析了遗传算法的搜索过程和效率问题 ,阐明了遗传算法作为一种次优算法的有效性 .相对于遗传算法的生物解释 ,本文给出了相应的物理解释 .同时 ,本文为遗传算法、进化策略和模拟退火算法找到了一种统一的物理解释 ,揭示了这些重要的仿生类算法实质上的相似性 .     

基于遗传算法的座位优化控制模型

座位优化控制是航空运输界增加利润的有效方法 .基于旅客的需求预测 ,可以利用数学规划模型为不同的航段和票价组合计算座位销售上限或者销售竞价 ,从而达到单个航班收入最大化的目的 .常用的方法可分为确定模型和概率模型 ,但对多航段多舱位的优化问题 ,由于出现了复杂的组合和约束 ,这些模型必须简化 .提出了基于遗传算法的座位优化控制模型 ,并和常用的优化方法进行了仿真对比 .研究结果表明 ,遗传算法应用于座位优化 ,可得到满意的解 ,同时 ,遗传算法简化了复杂的约束关系 ,易于实现 ,具有明显的优势 .     

信贷风险决策问题的一种相互逼近算法

基于最速下降法的基本思想 ,提出了相互逼近算法 ,用以解决信贷风险决策过程中 ,利润曲线和风险曲线寻求公共最优近似解的问题 .该算法表明 ,当利润曲线和风险曲线不存在公共最优近似解时 ,银行追求利润最大化的结果将导致风险上升 ,无法在可接受的风险指数范围内实现其既定的盈利目标 .但当利润曲线和风险曲线存在公共最优近似解时 ,银行根据其所掌握的私有信息以及所观测到企业理性的反应 ,作出相应的决策 .公共最优近似解的存在 ,说明了银行是在风险可接受的前提下按最优性原则给企业发放贷款 .     

二阶段随机规划问题基于随机模拟的遗传算法

利用遗传算法不过多依赖目标函数性质．适应于全局搜索的特点．提出了求解二阶段随机规划的基于随机模拟的遗传算法，算法采用随机模拟技术利用样本均值近似代替期望值，使计算得以简化，计算实例表明该算法是有效和可行的。     

Anisotropic mesh adaptation for numerical solution of boundary value problems

We present an efficient mesh adaptation algorithm that can be successfully applied to numerical solutions of a wide range of 2D problems of physics and engineering described by partial differential equations. We are interested in the numerical solution of a general boundary value problem discretized on triangular grids. We formulate a necessary condition for properties of the triangulation on which the discretization error is below the prescribed tolerance and control this necessary condition by the interpolation error. For a sufficiently smooth function, we recall the strategy how to construct the mesh on which the interpolation error is below the prescribed tolerance. Solving the boundary value problem we apply this strategy to the smoothed approximate solution. The novelty of the method lies in the smoothing procedure that, followed by the anisotropic mesh adaptation (AMA) algorithm, leads to the significant improvement of numerical results. We apply AMA to the numerical solution of an elliptic equation where the exact solution is known and demonstrate practical aspects of the adaptation procedure: how to control the ratio between the longest and the shortest edge of the triangulation and how to control the transition of the coarsest part of the mesh to the finest one if the two length scales of all the triangles are clearly different. An example of the use of AMA for the physically relevant numerical simulation of a geometrically challenging industrial problem (inviscid transonic flow around NACA0012 profile) is presented. © 2004 Wiley Periodicals, Inc. Numer Methods Partial Differential Eq, 2004.     

On the Convergence of a Population-Based Global Optimization Algorithm

In global optimization, a typical population-based stochastic search method works on a set of sample points from the feasible region. In this paper, we study a recently proposed method of this sort. The method utilizes an attraction-repulsion mechanism to move sample points toward optimality and is thus referred to as electromagnetism-like method (EM). The computational results showed that EM is robust in practice, so we further investigate the theoretical structure. After reviewing the original method, we present some necessary modifications for the convergence proof. We show that in the limit, the modified method converges to the vicinity of global optimum with probability one.     

A General Alternative Procedure for Solving Negative Degree of Difficulty Problems in Geometric Programming

The degree of difficulty is an important concept in classical geometric programming theory. The dual problem is often infeasible when the degree of difficulty is negative and little has been published on this topic. In this paper, an alternative procedure is developed to find the optimal solution for the posynomial geometric programming problem with a negative degree of difficulty. First an equivalent problem was constructed with a positive degree of difficulty and the general posynomial geometric programming problem was solved using an original method previously developed by the authors. This method avoids the difficulty of non-differentiability of the dual objective function in the classical methods classified as dual. It also avoids the problem that appears when the feasible region for the dual problem is formed by an inconsistent system of linear equations.