Showing papers on "Network planning and design published in 1985"

Quality Control in Geodetic Networks

Abstract: Generally one can consider the “quality of a network design” to be made up of three factors: Economy, Precision and Reliability Economy then expresses the costs of observation, transport, monumentation, etc; Precision as expressed by the a posteriori covariance matrix of the coordinates, is the measure of the network’s characteristics in propagating random errors and Reliability describes the ability of the redundant observations to check model errors

Numerical Methods in Network Design

Abstract: Although the history of computer aided optimisation of geodetic networks is rather short, probably less than twenty years, a vast number of different design strategies have been devised. All the practically useful methods share the disadvantage of needing to use expensive (in terms of computer storage and number of computer operations) numerical techniques in order to obtain the desired solution. Clearly the economy (if not the effectiveness) of any design procedure is largely dependent upon the numerical methods employed.

Aspects of Network Design

Abstract: Several aspects of network design are investigated in more detail, among them the datum problem for criterion matrices, the additional consideration of reliability constraints, the solution of design problems via the complementarity algorithm, and the transformation of the Third Order Design Problem for Gaus-Markov Models into a Second Order Design Problem for Mixed Models.

Analysis of traffic network flow relations and two-fluid model parameter sensitivity.

Abstract: Presented in this paper is a systematic exploration, using microscopic simulation, of the sensitivity of network-level traffic flow descriptors and relationships, particularly those of the two-fluid theory of town traffic, to network features, traffic control, and traffic-interfering urban activity levels. Moving traffic interference, which is represented by stochastic short-term lane blockages of varying duration and frequency, is shown to be a key determinant of the traffic character of an urban street network and of the behavior described by the two-fluid theory and verified operationally. In addition, the sensitivity of the two-fluid model parameters to a change in traffic control strategy, in this case the coordination of signals to achieve progression, is demonstrated. Furthermore, keeping the same network configuration, the effect of network topology on traffic flow is examined by changing the identical length of the links.

Properties of the iterative optimization-equilibrium algorithm

Abstract: This paper shows that the iterative optimization-equilibrium algorithm used to solve the continuous network design problem does not solve the original Abdulaal-LeBlanc normative network design model, but rather is an exact and efficient solution procedure for a descriptive model in which the planner and the network users compete in a Cournot-Nash setting. The existence of a solution to this descriptive model is proved, numerical results which illustrate convergence and efficiency of the algorithm are given, and suggestions for future research are made.

An integrated approach to cellular radio network planning

Abstract: The fundamental ideas behind a cellular network design program are described. By the aid of a generalization of the notion of a radio cell it becomes possible to integrate hitherto isolated areas of investigation. The elementary design steps making up the program are outlined together with their interconnections. The program may be used to design new networks, analyze or extend existing ones, and study system variants and design strategies in a general nonhomogeneous setting.

What is network planning

Abstract: The National Communications System (NCS) in support of its primary mission is exploring revolutionary communication concepts for the 21 century. These concepts are in the specialized fields of smart materials and structures, advanced network architectures, and intelligent systems. The purpose of this report is to assess the state-ofthe-art and state-of-the-practice in these technologies and in intelligent software agents related to these concepts. Further, the ability of the telecommunications infrastructure to support agent-based monitoring, detection, planning, cooperation, and decision-making in support of National Security and Emergency Preparedness (NS/EP) situations was also investigated. Specificity, the report focuses on the evolving network infrastructure, wireless networks and Internet-based networks, to support both intelligent systems and smart structures. The finding of this report suggest that the NCS should work with its governmental and industrial partners to promote NS/EP requirements in there specifications, designs and implementations of these revolutionary concepts so they can be seamlessly incorporated into legacy and future communications systems. In support of these efforts, NCS should create an environment that fosters the development of proof-of-concept experiments exploring the integration of these technologies in the context of NS/EP situations and scenarios; wherein, integrated intelligent systems incorporating smart structures could combine wireless and Internet technologies to allow devices to communicate over the Internet and Public Switched Telephone Network (PSTN).

Network Based Planning Techniques: A Critical Review of Published Developments

Abstract: Three areas of development of network planning are discussed; methods for handling uncertain activity durations, optimal planning when there are alternative time-cost options for some activities and planning when there are resource constraints. A review of published work in each area is given and illustrated by examples. An overview of the whole field is also presented.

Traffic models for telecommunication services with advance capacity reservation

Abstract: Certain new telecommunications services such as videoconferencing are based on the advance reservation of transmission channels. The corresponding traffic process is quite different to that of traditional telecommunications services and usual teletraffic models seem inadequate for system performance evaluation and network dimensioning. We discuss implied teletraffic issues and attempt to describe the reservation process by mathematical models. In particular we derive formulae useful for dimensioning a videoconference network to a low blocking probability grade of service standard.

Second-order design of combined linear-angular geodetic networks

Abstract: The paper deals with the solution of the weight problem for linear-angular networks on the base of criterion matrices. The observation plan of combined linear-angular networks consists of distances as well as of directions in any form. In particular the role of criterion matrices of completely isotropic structure for this type of two-dimensional network is discussed. Starting from an extreme network design, the results of a least-squares approximation are described analytically. In this case, the observation plan contains all geometrical point connections. Finally, the considerations are completed by giving as an example the second-order design of a control network for a dam of a water reservoir.

A Methodology for Optimal Network Design

Abstract: A methodology is developed to optimize the design of water distribution networks using reduced gradient techniques with a network simulator. The problem is reduced in complexity by incorporating a network simulator into the optimization model. This allows a more general problem to be solved including pipe sizing for new and expanding systems, pump sizing, valves, and storage facilities. In addition, the reduced problem can consider large networks under a number of loading conditions.

Hybrinet: a hybrid bus and token ring network

Abstract: We propose a hybrid local area network, called the Hybrinet, which combines all the good qualities of the CSMA/CD network and the token ring network. The performance of this network is shown to be superior to that of the token ring network for all offered loads where delay is finite. Furthermore, over the range of offered loads where the CSMA/CD network performs better than the token ring network, it is shown that the Hybrinet's performance is very close to that of the CSMA/CD network.

Network design and dynamic routing under queueing demand

Abstract: Given a nonhierarchical network and time-varying flow requirements, the problem of determining optimal capacities is termed design; that of determining optimal flows as dynamic routing. We formulate a linear program to solve both simultaneously in the case of deterministic flow requirements. A probability distribution termed the Erlang Difference Distribution is derived from a queueing model to describe random flow requirements, and this case leads to a separable convex program that has a linear programming equivalent. Both linear programs are amenable to Dantzig-Wolfe decomposition, which reveals subproblems that yield to special techniques of solution.

Probabilistic analysis of a network design problem heuristic

Abstract: Previous work has shown the budget network design problem (selecting a subset of arcs, subject to a budget constraint, so that the total weighted sum of the shortest paths in the network is minimized) to be a very difficult optimization problem. This article gives an efficient heuristic procedure for solving budget network design problems embedded in a circle on the euclidean plane where the nodes are independently and randomly distributed over the circle. With mild conditions on the budget constraint, we prove that as the number of nodes increases, the probability that the heuristic solution value exceeds any fixed percentage of the optimal solution value goes to zero. Computational results for the heuristic arc given.

A cost-effective multistage interconnection network with network overlapping and memory interleaving

Abstract: A cost-effective design is proposed for circuit-switching multistage interconnection networks (CSMINs). Increase of the network bandwidth and reduction of the network size (and thus low costs) are accomplished by network overlapping and memory interleaving (NOMI), instead of by increasing the number of switches or adding buffers. NOMI and its control principle are described on the basis of the structure and interconnection functions of CSMINs. Detailed accounts of both the network design and the drastic reduction in hardware costs are given. The impact of NOMI on system performance is also analyzed.

Augmenting Euclidean Networks—the Steiner Case

Abstract: On a Euclidean plane, a network and n points are given. It is required to interconnect the points and the network by links of minimal total length. The use of Steiner points is allowed, and connections can be made anywhere along the edges or to the vertices of the network. We prove that the problem can be solved in finite time by methods similar to those used for the Euclidean Steiner tree problem. The problem can be generalized to include flow dependent costs for the various links, or to allow for the connection of several networks. However, even in the form discussed in this paper, it may be useful for problems such as connecting new customers to existing networks (for example, computers, telecommunication, electricity, water, sewage disposal), especially if the projected flow between the points and the network does not justify more than the minimal possible investment.

Network Design with Multiple Demand: A New Approach

Abstract: The network design problem considered in this paper is an optimization problem, where a total demand cost function be minimized over budget constraints and network design variables, such as the topological structure and link capacities. The flow route selection, the separation of the flow in several commodities (With different source-sink couples), the indivisibility problems are not considered. The demand model proposed, take into account explicitly the uncertainty and the modifications of the demand vector during the operating period of the network. Therefore the demand cost function is a function of several feasible demand vectors instead of one. This network design problem with multiple demand is formulated as a linear program, Some properties of the model are analyzed and a solution procedure is proposed both for the general case and for some simple demand models.

An Optimal Road Network Design Model with Traffic Congestion and its Application

Abstract: An optimal road network design model is formulated as a two level planning problem. It is interpretted as a two-person, planner and user of road, non-cooperative non-zero sum game. Master problem decides continuous link capacities so as to minimize the sum of total transportation cost and total link construction cost subject to link capacity constraints. Sub problem is user equilibrium traffic assignment problem, in which traffic congestion is explicitly involved. A heuristic solution procedure is proposed which is effective for convex performance function and link construction cost function. In the case of BPR-type performance function and linear cost function, model application is executed for the actual size of road network planning problem.

A Strategy for Network Planning and Control

Abstract: Planning and implementing a data communications network is a complex task involving organizational as well as technical issues. The network design process must be integrated with the organization's business goals and strategies. Because network planning, implementation, and maintenance require skills not usually found in a batch-oriented DP environment, training is critical.

Integrated risk/cost planning models for the US Air Traffic system

Abstract: A prototype network planning model for the U.S. Air Traffic control system is described. The model encompasses the dual objectives of managing collision risks and transportation costs where traffic flows can be related to these objectives. The underlying structure is a network graph with nonseparable convex costs; the model is solved efficiently by capitalizing on its intrinsic characteristics. Two specialized algorithms for solving the resulting problems are described: (1) truncated Newton, and (2) simplicial decomposition. The feasibility of the approach is demonstrated using data collected from a control center in the Midwest. Computational results with different computer systems are presented, including a vector supercomputer (CRAY-XMP). The risk/cost model has two primary uses: (1) as a strategic planning tool using aggregate flight information, and (2) as an integrated operational system for forecasting congestion and monitoring (controlling) flow throughout the U.S. In the latter case, access to a supercomputer is required due to the model's enormous size.

Topology design and performance analysis of an integrated communication network

Abstract: A research study on the topology design and performance analysis for the Space Station Information System (SSIS) network is conducted It is begun with a survey of existing research efforts in network topology design Then a new approach for topology design is presented It uses an efficient algorithm to generate candidate network designs (consisting of subsets of the set of all network components) in increasing order of their total costs, and checks each design to see if it forms an acceptable network This technique gives the true cost-optimal network, and is particularly useful when the network has many constraints and not too many components The algorithm for generating subsets is described in detail, and various aspects of the overall design procedure are discussed Two more efficient versions of this algorithm (applicable in specific situations) are also given Next, two important aspects of network performance analysis: network reliability and message delays are discussed A new model is introduced to study the reliability of a network with dependent failures For message delays, a collection of formulas from existing research results is given to compute or estimate the delays of messages in a communication network without making the independence assumption The design algorithm coded in PASCAL is included as an appendix

APM Configuration Concepts and Network Design

Abstract: The simpler networks, along with several of the more complex variations, include the following: Single-Lane Shuttle, Single-Lane Shuttle with Bypass, Double-Lane Shuttle, Single-Lane Loop, Double-Lane Loop, Pinched Loop with Turnbacks, Multi-Shuttle, and Shuttle/Loop Combination. Given the numerous configurations, network planners must decide which concept best fits: the applications, and the performance requirements and design constraints associated with it. Several factors, and the relative importance placed on each of the factors by the planners, determine which network is best.

A methodology for the assessment of truck lane needs in the texas highway network. final report

Abstract: Special truck lanes have been proposed as a measure for dealing with the increasing traffic of larger and heavier trucks on the Texas highway system. This report describes an integrated network modelling methodology for the study of truck lane needs in the Texas highway network. It consists of three major components: critical link programming, network traffic assignment, and optimal link selection/network design. The critical link programs allow the user to diagnose and assess the adequacy of the links in a highway network for handling excessive truck traffic under specified conditions. The traffic assignment model is essential for the prediction of link flow patterns and subsequent user costs calculation in response to particular changes in the network corresponding to truck-related link improvements. The assignment problem addressed here allows the asymmetric interaction between cars and trucks sharing the roadway in the determination of link travel times. The optimal link improvement selection problem is cast as a discrete network design problem with multiple improvement options per link. One of its main features is the definition of link improvement in terms of both lane addition (capacity expansion) and operating scheme (lane access restrictions to cars and trucks).

Truck lane needs methodology: a heuristic approach to solve a five option discrete network design problem

Abstract: Special truck lanes have been proposed as a measure to deal with the increasing traffic of larger and heavier trucks on the Texas highway system. This report describes a procedure for the selection of an optimal subset of truck-related link improvements in the highway network. This procedure is a component of an integrated network modelling methodology for the study of truck lane needs in the Texas highway network. The link improvement selection problem is cast as a discrete network design problem with multiple improvement types per link. One of the principal features of this procedure is the definition of link improvement in terms of both capacity expansion (lane addition) and operational scheme (exclusive use by cars or trucks of both existing and added lanes). Another is the consideration of the interaction of cars and trucks in the traffic stream in solving the network equilibrium assignment problem embedded in the network design problem. A branch and bound integer programming approach is adapted and tested for this problem and the particular features introduced by truck-related link improvement measures.

Truck lane needs methodology: a heuristic approach to solve a five option discrete network design problem. interim report

Abstract: Special truck lanes have been proposed as a measure to deal with the increasing traffic of larger and heavier trucks on the Texas highway system. This report describes a procedure for the selection of an optimal subset of truck-related link improvements in the highway network. This procedure is a component of an integrated network modelling methodology for the study of truck lane needs in the Texas highway network. The link improvement selection problem is cast as a discrete network design problem with multiple improvement types per link. One of the principal features of this procedure is the definition of link improvement in terms of both capacity expansion (lane addition) and operational scheme (exclusive use by cars or trucks of both existing and added lanes). Another is the consideration of the interaction of cars and trucks in the traffic stream in solving the network equilibrium assignment problem embedded in the network design problem. A branch and bound integer programming approach is adapted and tested for this problem and the particular features introduced by truck-related link improvement measures.