Performance evaluation of artificial neural network approaches in forecasting reservoir inflow

This study investigates the potential of Time Lag Recurrent Neural Networks (TLRN) for modeling the daily inflow into Eleviyan reservoir, Iran. TLRN are extended with short term memory structures that have local recurrent connections, thus making them an appropriate model for processing temporal (time-varying) information. For this study, the daily inflow into Eleviyan reservoir between years 2004–2007 was considered. To compare the performance of TLRN, a back propagation neural network was used. The TLRN model with gamma memory structure, eight input layer nodes, two hidden layer and one output layer (8-2-1) was found performing best out of three different models used in forecasting daily inflow. A comparison of results with back propagation neural network suggest that neither TLRN nor back propagation approaches were good in forecasting high inflow but, both approaches perform well when used to forecast low inflow values. However, statistical test suggests that both TLRN and back propagation neural network models were able to reproduce similar basic statistics as that of the actual data.     

Impact of initial lattice structures on networks generated by traces of random walks

We show that the platform stage of network evolution plays a principal role in the topology of resulting networks generated by short-cuts stimulated by the movements of a random walker, the mechanism of which tends to produce power-law degree distributions. To examine the numerical results, we have developed a statistical method which relates the power-law exponent γ to random properties of the subgraph developed in the platform stage. As a result, we find that an important exponent in the network evolution is α, which characterizes the size of the subgraph in the form V∼t^α, where V and t denote the number of vertices in the subgraph and the time variable, respectively. 2D lattices can impose specific limitations on the walker’s diffusion, which keeps the value of α within a moderate range and provides typical properties of complex networks. 1D and 3D cases correspond to different ends of the spectrum for α, with 2D cases in between. Especially for 2D square lattices, a discontinuous change of the network structure is observed, which varies according to whether γ is greater or less than 2. For 1D cases, we show that emergence of nearly complete subgraphs is guaranteed by α<1/2, although the transient power-law is permitted at low increase rates of edges. Additionally, the model exhibits a spontaneous emergence of highly clustered structures regardless of its initial structure.     

A Graphical Tool to Visualize Predicted Minimum Delay Flights

In this article, we present a graphical decision support tool to aid in analyzing the U.S. air transport network. In addition to displaying simple statistics, our tool can calculate the predictions of both the minimum-delay and quickest routes for a given origin and destination airport using regression, simulation, and network optimization techniques. Using various visualizations allows for less obvious patterns in the data to be displayed. This article has supplementary material online.     

Efficiency decomposition for general multi-stage systems in data envelopment analysis

Conventional data envelopment analysis (DEA) models only consider the inputs supplied to the system and the outputs produced from the system in measuring efficiency, ignoring the operations of the internal processes. The results thus obtained sometimes are misleading. This paper discusses the efficiency measurement and decomposition of general multi-stage systems, where each stage consumes exogenous inputs and intermediate products (produced from the preceding stage) to produce exogenous outputs and intermediate products (for the succeeding stage to use). A relational model is developed to measure the system and stage efficiencies at the same time. By transforming the system into a series of parallel structures, the system efficiency is decomposed into the product of a modification of the stage efficiencies. Efficiency decomposition enables decision makers to identify the stages that cause the inefficiency of the system, and to effectively improve the performance of the system. An example of an electricity service system is used to explain the idea of efficiency decomposition.     

A Distributed and Collaborative Beamforming algorithm for a self-organizing Wireless Sensor Network

In this paper the scenario where sensors of a Wireless Sensor Network (WSN) are able to process and transmit monitored data to a far collector is considered. The far collector may be a Base Station (BS) that gathers data from a certain number of deployed WSNs, in applications such as earthquake, tsunami, or pollution monitoring. In this paper, the possible use of Distributed and Collaborative BeaMForming (DC-BMF) technique is analyzed, with the goal of enhancing the capability of a single sensor to communicate its data to the far collector. This technique considers nodes as elements of a phased array, where the phases of the signals at each antenna node are linearly combined in order to adjust the directional gain of the whole array. In particular, a novel self-localization technique for WSNs performing DC-BMF is studied, a closed form solution for beamforming gain degradation is derived and the evaluation of the power consumption of the proposed DC-BMF algorithm is provided.     

WDM Fiber Replacement Scheduling

We consider the migration of a WDM telecommunication network to a new technology. In the course of the migration process, shared network resources must be temporarily shut down, affecting the network connections that use them. In this paper we describe an ILP-based approach to find a migration schedule that minimizes the total service disruption occurring in the network.     

Transient fluctuation of the prosperity of firms in a network economy

The transient fluctuation of the prosperity of firms in a network economy is investigated with an abstract stochastic model. The model describes the profit which firms make when they sell materials to a firm which produces a product and the fixed cost expense to the firms to produce those materials and product. The formulas for this model are parallel to those for population dynamics. The swinging changes in the fluctuation in the transient state from the initial growth to the final steady state are the consequence of a topology-dependent time trial competition between the profitable interactions and expense. The firm in a sparse random network economy is more likely to go bankrupt than expected from the value of the limit of the fluctuation in the steady state, and there is a risk of failing to reach by far the less fluctuating steady state.     

A node rooted flow-based model for the local access network expansion problem

In this paper, we present a new formulation for the local access network expansion problem. Previously, we have shown that this problem can be seen as an extension of the well-known Capacitated Minimum Spanning Tree Problem and have presented and tested two flow-based models. By including additional information on the definition of the variables, we propose a new flow-based model that permits us to use effectively variable eliminations tests as well as coefficient reduction on some of the constraints. We present computational results for instances with up to 500 nodes in order to show the advantages of the new model in comparison with the others.     

On the L<Subscript>∞</Subscript>-norm of extreme points for crossing supermodular directed network LPs

We discuss extensions of Jain’s framework for network design [8] that go beyond undirected graphs. The main problem is approximating a minimum cost set of directed edges that covers a crossing supermodular function. We show that iterated rounding gives a factor 3 approximation, where factor 4 was previously known and factor 2 was conjectured. Our bound is tight for the simplest interpretation of iterated rounding. We also show that (the simplest version of) iterated rounding has unbounded approximation ratio when the problem is extended to mixed graphs.      

The multi-layered network design problem

We address the problem of designing a network built on several layers. This problem occurs in practical applications but has not been studied extensively from the point of view of global optimisation, since the problem of designing a single-layered network is complex. An example of an application is the design of a virtual network (Internet Protocol) built on a sparse optical transport network.