Effective Evolutionary Algorithms for Many-Specifications Attainment: Application to Air Traffic Control Tracking Filters

This paper addresses a real-world engineering design requiring the application of effective and global optimization techniques. The problem it deals with is the design of nonlinear tracking filters under up to several hundreds of performance specifications. The suitability of different evolutionary computation techniques for solving multiobjective problems is explored, contrasting the performance achieved with recent multiobjective evolutionary algorithm (MOEAs) proposals and different aggregation schemes. In particular, a new scheme is proposed to build a fitness function based on an operator that selects worst cases of multiple specifications in different situations. They have been evaluated in the design of an air traffic control (ATC) tracking filter that should accomplish a specific normative with 264 specifications. Results show their performance in terms of effectiveness and computational load, comparing their capability to scale the problem with respect to problem size.     

Visual evoked potentials generator model derived from different spatial frequency stimuli of visual field regions and magnetic resonance imaging coordinates of V1, V2, V3 areas in man

Visual evoked potentials (VEPs) to pattern reversal vertical bar stimuli were recorded from 24 scalp derivations (including zygomatic and inion) referenced to digitally linked earlobes in 50 controls. 1, 2 and 4 cpd patterns were presented as full field (FF) stimuli, on Upper Hemifields (UHF) and Lower Hemifields (LHF), upper and lower quadrants and with the occlusion of central and peripheral UHF and LHF. VEPs to octant stimuli were also recorded with 2 cpd patterns. N1, P1 and N2 components were recorded from posterior and inion derivations with FF stimuli, from posterior derivations with LHF stimuli, only from inion leads with UHF stimuli, from derivations ipsilateral to stimuli with quadrants and octants, and consistently from midline derivations only with lower quadrants. Polarity inverted sequences (iN1-iP1-iN2) were recorded from the other scalp derivations, with similar latency and spatial frequency sensitivity as N1-P1-N2. Single Equivalent Dipole (ED) calculations were performed on N1 and P1 recorded in the different stimulus conditions. Our findings contradict previous hypotheses on VEP generators and contradict the predictions of VEPs polarity and distribution based on the "cruciform model" of VEPs generators. In order to explain the distribution of VEPs to upper and lower half fields and to quadrant and octants, we propose a model based on the position of the medial and occipito-polar surface of visual cortex in man.     

Flexural behavior of stratified reinforced concrete: construction,testing, analysis,and design

Stratified concrete poses a promising alternative for construction. Its fresh and hardened properties have been studied at the material level; however, structural behavior in steel reinforced specimens has not been studied. This paper focuses on the flexural behavior of eight stratified reinforced concrete (SRC) specimens representing slices from a slab or non-bearing wall. Specimens with two stratified concrete designs and three steel ratios were tested and compared to estimates from a fiber element numerical model and rectangular stress-block design methods from ACI 318 and Eurocode 2. The results suggest that SRC has similar damage modes as ordinary reinforced concrete (ORC). The fiber element model accurately estimated the measured behavior, while ACI 318 and Eurocode 2 differed from the experimental results by <25%. These prediction accuracies are similar to those for ORC. Therefore, the flexural design of SRC can be done using both fiber element and rectangular stress-block approaches.     

Flow modeling of linear and nonlinear fluids in two scale fibrous fabrics

The fundamental macroscopic material property needed to quantify the flow in a fibrous medium viewed as a porous medium is the permeability. Composite processing models require the permeability as input data to predict flow patterns and pressure fields. As permeability reflects both the magnitude and anisotropy of the fluid/fiber resistance, efficient numerical techniques are needed to solve linear and nonlinear homogenization problems online during the flow simulation. In a previous work the expressions of macroscopic permeability were derived in a double-scale porosity medium for both Newtonian and rheo-thinning resins. In the linear case only a microscopic calculation on a representative volume is required, implying as many microscopic calculations as representative microscopic volumes exist in the whole fibrous structure. In the non-linear case, and even when the porous microstructure can be described by a unique representative volume, microscopic calculation must be carried out many times because the microscale resin viscosity depends on the macroscopic velocity, which in turn depends on the permeability that results from a microscopic calculation. Thus, a nonlinear multi-scale problem results. In this paper an original and efficient offline-online procedure is proposed for the efficient solution of nonlinear flow problems in porous media.     

Innovative techniques in pediatric liver transplantation: reduced-, split- and living-donor related liver transplantation

BACKGROUND: Biological systems that show threshold phenomena are candidates for stochastic resonance as a mechanism to explain what appears to be biovariability. Stochastic resonance is the amplification of weak signals by "noise." OBJECTIVE: This review discusses the areas of contact dermatitis in which threshold events have been documented. The purpose is to point out the mechanism by which stochastic resonance may affect patch test results. METHODS: A literature review technique was used. RESULTS: The recent finding of a neurological influence on contract dermatitis provides a mechanism for stochastic resonance to affect patch test results. CONCLUSION: Stochastic resonance is likely to affect patch test results when more than one patch test result is positive.     

Discrepancy search for the flexible job shop scheduling problem

The flexible job shop scheduling problem (FJSP) is a generalization of the classical job shop problem in which each operation must be processed on a given machine chosen among a finite subset of candidate machines. The aim is to find an allocation for each operation and to define the sequence of operations on each machine, so that the resulting schedule has a minimal completion time. We propose a variant of the climbing discrepancy search approach for solving this problem. We also present various neighborhood structures related to assignment and sequencing problems. We report the results of extensive computational experiments carried out on well-known benchmarks for flexible job shop scheduling. The results demonstrate that the proposed approach outperforms the best-known algorithms for the FJSP on some types of benchmarks and remains comparable with them on other ones.     

Monte Carlo techniques for phrase-based translation

Recent advances in statistical machine translation have used approximate beam search for NP-complete inference within probabilistic translation models. We present an alternative approach of sampling from the posterior distribution defined by a translation model. We define a novel Gibbs sampler for sampling translations given a source sentence and show that it effectively explores this posterior distribution. In doing so we overcome the limitations of heuristic beam search and obtain theoretically sound solutions to inference problems such as finding the maximum probability translation and minimum risk training and decoding.