A generic auction algorithm for the minimum cost network flow problem

In this paper we broadly generalize the assignment auction algorithm to solve linear minimum cost network flow problems. We introduce a generic algorithm, which contains as special cases a number of known algorithms, including the -relaxation method, and the auction algorithm for assignment and for transportation problems. The generic algorithm can serve as a broadly useful framework for the development and the complexity analysis of specialized auction algorithms that exploit the structure of particular network problems. Using this framework, we develop and analyze two new algorithms, an algorithm for general minimum cost flow problems, called network auction, and an algorithm for thek node-disjoint shortest path problem.     

One-parametric bottleneck transportation problems

We discuss the bottleneck transportation problem with one nonlinear parameter in the bottleneck objective function. A finite sequence of feasible basic solutions which are optimal in subsequent closed parameter-intervals is generated using a primal method for the nonparametric subproblems. The best among three primal codes for solving these subproblems is selected on extensive computational comparisons. We discuss computational experience with the sequential method for the case of linear and quadratic dependence on one parameter. Observed computational behaviour is O((n ·m)), with 2.     

An atom/molecule/DNA probing and transportation using dynamic optical tweezers via a wavelength router

A dark soliton pulse beam that has potential applications in the probing and transport of atoms or molecules as an optical tweezers is demonstrated. The trapped pulse beam is formed in the gap between a BP signal at 1500.00 nm and Stokes signal at 1500.09 nm, which has peak powers of −12 dBm. The generated beam is seen to be stable with no fluctuations over a test period of 10 min, which is important to ensure that the transported atom/molecule is not lost in the link media, therefore, the dynamic optical tweezers are now plausible. The theoretical background of the trapped atom/molecule is also analyzed and described in details. The atomic/molecular network via a wavelength router is also described.     

Mass Transport and Variants of the Logarithmic Sobolev Inequality

We develop the optimal transportation approach to modified log-Sobolev inequalities and to isoperimetric inequalities. Various sufficient conditions for such inequalities are given. Some of them are new even in the classical log-Sobolev case. The idea behind many of these conditions is that measures with a non-convex potential may enjoy such functional inequalities provided they have a strong integrability property that balances the lack of convexity. In addition, several known criteria are recovered in a simple unified way by transportation methods and generalized to the Riemannian setting. The research of A.V. Kolesnikov was supported by RFBR 07-01-00536, DFG Grant 436 RUS 113/343/0 and GFEN 06-01-39003.     

一种新的交通网络设计优化算法

交通网络设计问题是研究如何用定量的方法在已有交通网络上添加或扩容某些路段的问题.文章在回顾交通网络设计问题文献的基础上,提出了基于图论网络优化思想的解决该类问题的一种新思路,给出了启发式算法,并进行了算法复杂性分析,最后通过算例验证了其有效性.     

基于Stackelberg博弈的时变双层交通分配模型

提出一个时变双层交通分配模型,其中上层网络管理者设立了一个路段的最大排队长度,其目标是使由网络流和排队长度定义的总出行时间最小.目标函数在离散时段内以路段流量和排队长度作为决策变量,同时考虑不同类型的信号交叉口延误的影响.下层网络用户的反应依赖于上层管理者的决策,其选择是使自身感知阻抗最小的路径,服从一个基于成对组合Logit的路径选择模型,构成一个成对组合Logit的均衡分配问题.结合了交通分配和流传播方法,将其表示为一个均衡约束下的双层数学规划问题,形成了一个Stackelberg非合作博弈.使用遗传算法求解该双层规划问题,并采用实证分析来表现模型的特征和算法的计算表现.结果表明路径重叠、路段流量、路段排队长度等因素对网络均衡流分布均有显著影响.     

基于深度学习LSTM对交通流状态的预测

当今道路交通状态对城市管理和人们出行愈加重要,影响着人类生活的方方面面.以深圳交通为研究对象,由基础车辆数据和道路坐标构建了路网系统,从车辆速度和密度两个方面导出了交通流状态评价指数TSI.利用深度学习长短期记忆神经网络(LSTM)对车辆速度和密度两个指标进行预测,并通过对比极限学习机(ELM),时间序列(ARMA)和BP神经网络,进行仿真实验,结果表明相对于传统预测模型,所采用的LSTM网络具有更优的预测精确度和对远期预测的稳定性.最后利用预测结果计算出更能直观反映出道路交通拥堵情况的TSI指数,为人们提供了准确的交通状态预测.     

Brascamp-Lieb inequality for positive double John basis and its reverse

In this paper, we establish the Brascamp-Lieb inequality for positive double John basis and its reverse. As their applications, we estimate the upper and lower bounds for the volume product of two unit balls with the given norms. Moreover, the Loomis-Whitney inequality for positive double John basis is obtained.     

Ship routing and scheduling in the new millennium

We review research on ship routing and scheduling and related problems during the new millennium and provide four basic models in this domain. The volume of research in this area about doubles every decade as does the number of research outlets. We have found over a hundred new refereed papers on this topic during the last decade. Problems of wider scope have been addressed as well as more specialized ones. However, complex critical problems remain wide open and provide challenging opportunities for future research.