The particulate matter PM10 concentrations have been impacting hospital admissions due to respiratory diseases. The air pollution studies seek to understand how this pollutant affects the health system. Since prediction involves several variables, any disparity causes a disturbance in the overall system, increasing the difficulty of the models’ development. Due to the complex nonlinear behavior of the problem and their influencing factors, Artificial Neural Networks are attractive approaches for solving estimations problems. This paper explores two neural network architectures denoted unorganized machines: the echo state networks and the extreme learning machines. Beyond the standard forms, models variations are also proposed: the regularization parameter (RP) to increase the generalization capability, and the Volterra filter to explore nonlinear patterns of the hidden layers. To evaluate the proposed models’ performance for the hospital admissions estimation by respiratory diseases, three cities of São Paulo state, Brazil: Cubatão, Campinas and São Paulo, are investigated. Numerical results show the standard models’ superior performance for most scenarios. Nevertheless, considering divergent intensity in hospital admissions, the RP models present the best results in terms of data dispersion. Finally, an overall analysis highlights the models’ efficiency to assist the hospital admissions management during high air pollution episodes.

A relevant research topic in optimization of power distribution networks is to find the best relationship between system reliability and the allocation of maintenance resources. This paper presents a mathematical formulation that seeks the optimal preventive maintenance budget regarding the system reliability constraints. A dynamic programming approach is proposed to deal with this optimization problem. Some reductions are applied to the dynamic programming approach in order to avoid the combinatorial explosion. Case studies are presented to compare the performance of the dynamic programming approach with a hybrid genetic algorithm previously developed.

No abstract

Foraminifers are widespread, highly abundant protists and active participants in marine carbon cycling. Their biomass might represent almost half of the total meiobenthic biomass in the deep sea. Foraminiferal biomass is frequently assessed through geometric models and biovolume estimates due to its non-destructive nature, which allows estimates of individuals from palaeoecological, museum, and living samples. To increase the accuracy of foraminiferal biovolume and biomass assessment we evaluate and propose geometric models for 207 foraminiferal taxa and the species’ average cell occupancy of the test. Individual test dimensions were measured to calculate volume (µm³), and the percent of cell occupancy (PCO) of the test was measured to assess the biovolume (µm³). These data were converted into individual biomass measurements (µg Corg ind−1). Our high intra- and interspecific PCO variance suggest that a mean PCO for each species represents the natural variability of occupancy more accurately than a predetermined fixed percentage for the whole assemblage, as previously asserted in the literature. Regression equations based on the relationship between test dimensions and volumes are presented. The geometric models, the PCO adjustment, and the equations will reduce time, effort, and discrepancies in foraminiferal biovolume and biomass assessments. Therefore, these results can improve the use and reliability of foraminiferal biomass in the future, facilitating its use in (1) distinct approaches including carbon flux estimations, (2) determining the effects of climate change on the marine trophic webs, and (3) environmental monitoring programs.

Palavras-chave: otimização de manutenções, manutenção baseada em confiabilidade, redes de distribuição de energia elétrica, meta-heurísticas híbridas, buscas em espaço de estados, Simulated Annealing.