Critical Thresholds for Dynamic Routing in Queueing Networks

This paper studies dynamic routing in a parallel server queueing network with a single Poisson arrival process and two servers with exponential processing times of different rates. Each customer must be routed at the time of arrival to one of the two queues in the network. We establish that this system operating under a threshold policy can be well approximated by a one-dimensional reflected Brownian motion when the arrival rate to the network is close to the processing capacity of the two servers. As the heavy traffic limit is approached, thresholds which grow at a logarithmic rate are critical in determining the behavior of the limiting system. We provide necessary and sufficient conditions on the growth rate of the threshold for (i) approximation of the network by a reflected Brownian motion (ii) positive recurrence of the limiting Brownian diffusion and (iii) asymptotic optimality of the threshold policy.     

A New Multiobjective Dynamic Routing Method for Multiservice Networks: Modelling and Performance

There are potential advantages in formulating the routing problems in modern multiservice networks as multiple objective problems. This paper presents a novel hierarchical bi-level multiobjective dynamic routing model for multiservice networks. It is based on a bi-objective shortest path algorithm, with dynamically adapted soft-constraints, to compute alternative paths for each node pair and on a heuristic to synchronously select alternative routing plans for the network in a dynamic alternative routing context. It is a routing method which periodically changes alternative paths as a function of periodic updates of certain QoS related parameters obtained from real-time measurements. The performance of the proposed routing method is compared with two reference dynamic routing methods namely RTNR and DAR by means of a discrete-event simulator.A previous short version of this work was presented at INOC’03 (International Network Optimisation Conference). Work partially supported by programme POSI of the III EC programme cosponsored by FEDER and national funds.     

Bandwidth allocation in ATM networks

In an ATM network, bandwidth is allocated at different levels and in different stages. At the physical level, the ATM topology can be dynamically reconfigured by adding/removing truns between ATM switches. This allocation of bandwidth is made possible by the SONET Synchronous Transfer Mode (STM) infrastructure equipped with Digital Cross Connect Systems (DCSs). We will refer to this allocation asSTM allocation. At the ATM level, we can allocate bandwidth to individual Virtual Circuits (ATM-VC allocation) as well as to Virtual Paths (ATM-VP allocation). For example, in order to implement the Connectionless Network Access layer functions we find it convenient to organize the Virtual Paths in a Connectionless Overlay Network. This introduces another type of bandwidth allocation (CLS allocation). In this paper, we address and formulate the above bandwidth allocation problems, and propose efficient techniques for their solution. We illustrate these techniques with examples based on STM and CLS allocation, respectively.This work was jointly supported by the Brazilian National Science Council (CNPq) and NSF.     

Ad Hoc网络中基于时分的分布式QoS路由算法

捷径冲突是AdHvc网络中QoS路由特有的一种现象，它由WenjianShao在[2]中首次提出．本文进一步研究了捷径冲突现象，给出了一个更加准确的定义，且给出了一个基于时分的分布式QoS路由算法成功地避免了捷径冲突现象．本算法是基于TDMA的分布式算法，每个节点只需了解网络的局部信息即可．数据分析表明本算法预留的最大带宽比较接近AdHoc网络中所能用的最大带宽．     

车辆路径问题的改进遗传算法

提出一种基于遗传算法的求解车辆路径问题的新算法,避免传统遗传算法处理不可行约束条件中惩罚项系数选取不当所出现的问题.同时,通过现实例子分析该算法的优劣性,实验结果表明该算法是一种有效的算法.     

一类网络速率控制和路由联合算法的稳定性

近年来,动态多路径路由下网络速率控制的研究受到广泛关注.本文提出了一个新的速率控制和多路径路由联合的算法,该算法的特点是具有唯一的平衡点.利用传统的Lyapunov方法,我们证明算法在没有传播时延情形下的全局稳定性.而且,更为重要的是,即使考虑传播时延,在一定的条件下,该算法是局部稳定的.在平衡点处,每条路由上的速率非零.这一事实不但去掉了Kelly F P,Voice T(2005)结果中内部平衡点的假设条件,而且也可以理解为一种探测机制.我们通过仿真证实了算法的正确性,同时仿真结果也表明局部稳定性的吸引域可以很大,甚至是全局稳定的.     

Layouts for mobility management in wireless ATM networks

In this paper, we present a new model that combines quality of service and mobility aspects in wireless ATM networks. Namely, besides the hop count and load parameters of the basic ATM layouts, we introduce a new notion of distance that estimates the time needed to reconstruct the virtual channel of a wireless user when he moves through the network. Quality of service guarantee dictates that the rerouting phase must be imperceptible, that is, the maximum distance between two virtual channels must be maintained as low as possible. Therefore, a natural combinatorial problem arises in which suitable trade-offs must be determined between the different performance measures. We first show that establishing the existence of a layout with maximum hop count h, load l and distance d is NP-complete, even in the very restricted case h=2, l=1 and d=1. We then provide optimal layout constructions for basic interconnection networks, such as chains and rings.     

立方体网络路由选择算法

本文利用图论理论 ,基于路由选择能力的概念 ,建立了一个有效的路由选择算法 ,该算法可以在含有节点故障和边故障的容错超立方体上使用 ,且具有较强的容错性 .     

相异路径选择问题的模型与仿真结果研究

相异路径选线问题考虑的是,给定一运输网络的源宿接点,找出源宿节点之间在空间上有差异的路径,多个相异路径对于军事后勤供应,有害物品运输等在异常情况（如原来的最佳线路因气候等原因不可用）下的决策,具有重要意义。本着重对四种生成空间相异路径的算法进行了分析和评价,并构建了基于边的通道最短路的相异路径的算法,在此基础上,主要对IRM,GSP和EGSP给出了仿真实验计算结果,通过对结果的分析,找到了影响结果的几个关键因素。     

Virtual Path Design in Service-Specific ATM Networks

This paper addresses the problem of virtual path management in ATM networks, which is the problem of jointly selecting efficient virtual trunk routes and sizing them to meet end-to-end grade-of-service requirements. The problem is posed over capacitated networks and is formulated as a two-level multi-commodity network flow problem with linear side-constraints (physical layer capacity) and non-linear side constraints (end-to-end/link blocking). Through a variety of examples we show the method (i) generates solutions that agree with engineering judgement, (ii) can solve VP layout management for realistic size networks (of up to 200 nodes) in reasonable time and (iii) provides upper bounds on how far the solution strays from the mathematically optimal design.     

Evolutionary Algorithms for the Vehicle Routing Problem with Time Windows

This paper surveys the research on evolutionary algorithms for the Vehicle Routing Problem with Time Windows (VRPTW). The VRPTW can be described as the problem of designing least cost routes from a single depot to a set of geographically scattered points. The routes must be designed in such a way that each point is visited only once by exactly one vehicle within a given time interval. All routes start and end at the depot, and the total demands of all points on one particular route must not exceed the capacity of the vehicle. The main types of evolutionary algorithms for the VRPTW are genetic algorithms and evolution strategies. In addition to describing the basic features of each method, experimental results for the benchmark test problems of Solomon (1987) and Gehring and Homberger (1999) are presented and analyzed.     

Optimal Control for Traffic Flow Networks

We consider traffic flow models for road networks where the flow is controlled at the nodes of the network. For the analytical and numerical optimization of the control, the knowledge of the gradient of the objective functional is useful. The adjoint calculus introduced below determines the gradient in two ways. We derive the adjoint equations for the continuous traffic flow network model and derive also the adjoint equations for a discretized model. Numerical examples for the solution of problems of optimal control for traffic flow networks are presented.This author was supported by Deutsche Forschungsgemeinschaft (DFG), Grant KL 1105/5.     

A Queueing Framework for Routing Problems with Time-dependent Travel Times

Assigning and scheduling vehicle routes in a dynamic environment is a crucial management problem. Despite numerous publications dealing with efficient scheduling methods for vehicle routing, very few addressed the inherent stochastic and dynamic nature of travel times. In this paper, a vehicle routing problem with time-dependent travel times due to potential traffic congestion is considered. The approach developed introduces the traffic congestion component based on queueing theory. This is an innovative modelling scheme to capture the stochastic behavior of travel times as it generates an analytical expression for the expected travel times as well as for the variance of the travel times. Routing solutions that perform well in the face of the extra complications due to congestion are developed. These more realistic solutions have the potential to reduce real operating costs for a broad range of industries which daily face routing problems. A number of datasets are used to illustrate the appropriateness of the novel approach. Moreover it is shown that static (or time-independent) solutions are often infeasible within a congested traffic environment which is generally the case on European road networks. Finally, the effect of travel time variability (obtained via the queueing approach) is quantified for the different datasets.      

LP-Based Heuristic Algorithms for Interconnecting Token Rings via Source Routing Bridges

We develop a method to determine the topology of a network that interconnects a number of token rings using source routing bridges. The purpose is to compute a topology that provides low response delays for network users at a minimal cost of bridge installations. We formulate this network design problem as a mixed binary integer linear program. We develop effective heuristic algorithms. The algorithms exploit the topology and routing solutions of the linear programming relaxation in a sophisticated manner which we believe is new in the literature. The model incorporates performance issues, such as network stability, bridge overflow, back pressure effect and broadcast storm, that are specific to the underlying communication technology. By formally incorporating these performance issues, we tighten the model formulation and improve the quality of the LP bound considerably. Computational results are reported for problems with up to 20 token rings and 190 potential bridge locations.     

A Neural-Tabu Search Heuristic for the Real Time Vehicle Routing Problem

The Real Time Vehicle Routing Problem RTVRP is a dynamic routing problem where requests are generated dynamically during the operation horizon without any previous knowledge. Received requests need to be answered as fast as possible and then assigned to a vehicle to be served. Due to timing constraints of the RTVRP, a solving approach should give the best compromise between the cost of the provided solution and the computation time needed to find it. In this paper, we present a neural-tabu search solving scheme for the RTVRP. The developed approach is composed by two phases; The first part consists of learning and reproducing previous routing decisions using a feed forward neural network with a particular structure. The second phase is based on a tabu search heuristic that takes its initial solution from the assignment provided by the neural module. If the reaction time is still available, the tabu search module will continue ameliorating the final solution. To evaluate the proposed approach a set of problems are simulated and solved. The obtained results are compared to those given by the First Come First Served FCFS and Nearest Neighbor NN policies and also to the optimal solutions provided by the GNU Linear Programming Kit GLPK.      

几种特殊结构全光网中多播路由的波长分配算法

研究全光WDM网络中多播请求的路由与波长分配问题.给定网络拓扑和一组多播通信请求,要求对其进行路由和波长分配,满足波长连续性和波长无冲突约束,使得所用的波长总数最少.就几类特殊网络进行了研究.首先对二分树网络进行了研究,此时问题是多项式时间可求解的.其次对树网络进行了讨论,证明了即使是星网络,问题也不存在近似比小于m1/2-ρ(0＜ρ＜1))的近似算法,除非NP=ZPP,这里m是星图的边数.随后给出了近似比为(√m 1)(log r/√m 1 1)的近似算法,此结果对一般图也成立.最后考虑了环网和树环网,给出了近似比为3.6和2△的近似算法,这里△是图的最大度.     

双需求集货送货一体化车辆路径问题的数学模型及算法

本文研究了具有总时间和车容量约束的双需求集货送货一体化车辆路径问题,在综合考虑运输费用和车辆出行固定费用的前提下,建立了该问题的整数线性规划模型,并分别给出了求解该模型的精确算法和基于节约准则的启发式算法。最后通过一个具体实例验证了算法的有效性。     

基于聚类和动态规划的组合路径策略

为提高电商仓库的拣货作业效率,本文提出了基于聚类和动态规划的组合路径策略,实现了生成路径消耗时间和路径长度之间的平衡,并将这一策略成功地应用到多区型仓库。该策略分四步:首先,根据待拣储位分布特征,运用聚类分析法对其进行分类;然后,以各类的首末储位作为节点,运用动态规划法对已得分类进行排序,得到相应的类序;其次,得到各类内部路径;最后,依次拣取待拣商品,并返回出发点完成拣货作业。在提出新的路径策略后,通过仿真方法将新策略与三种传统路径策略(穿越策略、最大间隙策略和混合策略)和一种优质算法(蚁群算法)进行了对比分析,结果表明:该策略具备良好的适用性和实用性。     

WDM网络中的一个改进的最优半光通道路由算法

本文在一个限定的条件下，提出了一个WDM网络中的寻找最优半光通道算法，使时间复杂度从O（k^2n km knlog(kn))提高到O（k^2n km nlogn)。     

利用动态贝叶斯网构建基因调控网络的研究进展

构建基因调控网络是21世纪人类科学所面临的重要挑战之一。基因调控网络是一个基因组内基因相互作用而形成的关系网络,它从全基因组水平上以系统和全局的角度来研究复杂的生命现象及其本质。本文阐述了近几年来此领域的研究进展,着重介绍利用动态贝叶斯网络重构基因调控网络的若干模型,包括加权核l1模型,正则化模型、高斯混合贝叶斯网模型和自回归时间变化模型。