Efficient knowledge retrieval to calibrate input variables in forest fire prediction

Forest fires are a serious threat to humans and nature from an ecological, social and economic point of view. Predicting their behaviour by simulation still delivers unreliable results and remains a challenging task. Latest approaches try to calibrate input variables, often tainted with imprecision, using optimisation techniques like Genetic Algorithms. To converge faster towards fitter solutions, the GA is guided with knowledge obtained from historical or synthetical fires. We developed a robust and efficient knowledge storage and retrieval method. Nearest neighbour search is applied to find the fire configuration from knowledge base most similar to the current configuration. Therefore, a distance measure was elaborated and implemented in several ways. Experiments show the performance of the different implementations regarding occupied storage and retrieval time with overly satisfactory results.

Analysis of the effects of conical indentation variables on the indentation response of elastic-plastic materials through factorial design of experiment

In this work, the effects of conical indentation variables on the load-depth indentation curves were analyzed using finite element modeling and dimensional analysis. A factorial design 2(6) was used with the aim of quantifying the effects of the mechanical properties of the indented material and of the indenter geometry. Analysis was based on the input variables Y/E, R/h(max), n, theta, E, and h(max). The dimensional variables E and h(max) were used such that each value of dimensionless Y/E was obtained with two different values of E and each value of dimensionless R/h(max) was obtained with two different h(max) values. A set of dimensionless functions was defined to analyze the effect of the input variables: Pi(1) = P(1)/Eh(2), Pi(2) = h(c)/h, Pi(3) = H/Y, Pi(4) = S/Eh(max), Pi(6) = h(max)/h(f) and Pi(7) = W(P)/W(T). These six functions were found to depend only on the dimensionless variables studied (Y/E, R/h(max), n, theta). Another dimension less function, Pi(5) = beta, was not well defined for most of the dimensionless variables and the only variable that provided a significant effect on beta was theta. However, beta showed a strong dependence on the fraction of the data selected to fit the unloading curve, which means that beta is especially Susceptible to the error in the Calculation of the initial unloading slope.

On the utility of input selection and pruning for financial distress prediction models

Analyzes the use of linear and neural network models for financial distress classification, with emphasis on the issues of input variable selection and model pruning. A data-driven method for selecting input variables (financial ratios, in this case) is proposed. A case study involving 60 British firms in the period 1997-2000 is used for illustration. It is shown that the use of the Optimal Brain Damage pruning technique can considerably improve the generalization ability of a neural model. Moreover, the set of financial ratios obtained with the proposed selection procedure is shown to be an appropriate alternative to the ratios usually employed by practitioners.

An Input-Process-Output Approach to Interorganizational Teams: The Influence of Work Group Diversity, Trust and Shared Leadership on Communication Network and Team Outputs.

The market’s challenges bring firms to collaborate with other organizations in order to create Joint Ventures, Alliances and Consortia that are defined as “Interorganizational Networks” (IONs) (Provan, Fish and Sydow; 2007). Some of these IONs are managed through a shared partecipant governance (Provan and Kenis, 2008): a team composed by entrepreneurs and/or directors of each firm of an ION. The research is focused on these kind of management teams and it is based on an input-process-output model: some input variables (work group’s diversity, intra-team's friendship network density) have a direct influence on the process (team identification, shared leadership, interorganizational trust, team trust and intra-team's communication network density), which influence some team outputs, individual innovation behaviors and team effectiveness (team performance, work group satisfaction and ION affective commitment). Data was collected on a sample of 101 entrepreneurs grouped in 28 ION’s government teams and the research hypotheses are tested trough the path analysis and the multilevel models. As expected trust in team and shared leadership are positively and directly related to team effectiveness while team identification and interorganizational trust are indirectly related to the team outputs. The friendship network density among the team’s members has got positive effects on the trust in team and on the communication network density, and also, through the communication network density it improves the level of the teammates ION affective commitment. The shared leadership and its effects on the team effectiveness are fostered from higher level of team identification and weakened from higher level of work group diversity, specifically gender diversity. Finally, the communication network density and shared leadership at the individual level are related to the frequency of individual innovative behaviors. The dissertation’s results give a wider and more precise indication about the management of interfirm network through “shared” form of governance.

Regression with input-dependent noise: A Bayesian treatment

In most treatments of the regression problem it is assumed that the distribution of target data can be described by a deterministic function of the inputs, together with additive Gaussian noise having constant variance. The use of maximum likelihood to train such models then corresponds to the minimization of a sum-of-squares error function. In many applications a more realistic model would allow the noise variance itself to depend on the input variables. However, the use of maximum likelihood to train such models would give highly biased results. In this paper we show how a Bayesian treatment can allow for an input-dependent variance while overcoming the bias of maximum likelihood.

Input/output deterioration in production processes

In many real applications of Data Envelopment Analysis (DEA), the decision makers have to deteriorate some inputs and some outputs. This could be because of limitation of funds available. This paper proposes a new DEA-based approach to determine highest possible reduction in the concern input variables and lowest possible deterioration in the concern output variables without reducing the efficiency in any DMU. A numerical example is used to illustrate the problem. An application in banking sector with limitation of IT investment shows the usefulness of the proposed method. © 2010 Elsevier Ltd. All rights reserved.

Trained Neural Network Characterizing Variables for Predicting Organic Retention by Nanofiltration Membranes

Many organic compounds cause an irreversible damage to human health and the ecosystem and are present in water resources. Among these hazard substances, phenolic compounds play an important role on the actual contamination. Utilization of membrane technology is increasing exponentially in drinking water production and waste water treatment. The removal of organic compounds by nanofiltration membranes is characterized not only by molecular sieving effects but also by membrane-solute interactions. Influence of the sieving parameters (molecular weight and molecular diameter) and the physicochemical interactions (dissociation constant and molecular hydrophobicity) on the membrane rejection of the organic solutes were studied. The molecular hydrophobicity is expressed as logarithm of octanol-water partition coefficient. This paper proposes a method used that can be used for symbolic knowledge extraction from a trained neural network, once they have been trained with the desired performance and is based on detect the more important variables in problems where exist multicolineality among the input variables.

Comparison between detailed model simulation and artificial neural network for forecasting building energy consumption

There are several ways to attempt to model a building and its heat gains from external sources as well as internal ones in order to evaluate a proper operation, audit retrofit actions, and forecast energy consumption. Different techniques, varying from simple regression to models that are based on physical principles, can be used for simulation. A frequent hypothesis for all these models is that the input variables should be based on realistic data when they are available, otherwise the evaluation of energy consumption might be highly under or over estimated. In this paper, a comparison is made between a simple model based on artificial neural network (ANN) and a model that is based on physical principles (EnergyPlus) as an auditing and predicting tool in order to forecast building energy consumption. The Administration Building of the University of Sao Paulo is used as a case study. The building energy consumption profiles are collected as well as the campus meteorological data. Results show that both models are suitable for energy consumption forecast. Additionally, a parametric analysis is carried out for the considered building on EnergyPlus in order to evaluate the influence of several parameters such as the building profile occupation and weather data on such forecasting. (C) 2008 Elsevier B.V. All rights reserved.

MODELING ELECTRODIALYSIS AND A PHOTOCHEMICAL PROCESS FOR THEIR INTEGRATION IN SALINE WASTEWATER TREATMENT

Oxidation processes can be used to treat industrial wastewater containing non-biodegradable organic compounds. However, the presence of dissolved salts may inhibit or retard the treatment process. In this study, wastewater desalination by electrodialysis (ED) associated with an advanced oxidation process (photo-Fenton) was applied to an aqueous NaCl solution containing phenol. The influence of process variables on the demineralization factor was investigated for ED in pilot scale and a correlation was obtained between the phenol, salt and water fluxes with the driving force. The oxidation process was investigated in a laboratory batch reactor and a model based on artificial neural networks was developed by fitting the experimental data describing the reaction rate as a function of the input variables. With the experimental parameters of both processes, a dynamic model was developed for ED and a continuous model, using a plug flow reactor approach, for the oxidation process. Finally, the hybrid model simulation could validate different scenarios of the integrated system and can be used for process optimization.

MPC for tracking zone regions

This paper deals with the problem of tracking target sets using a model predictive control (MPC) law. Some MPC applications require a control strategy in which some system outputs are controlled within specified ranges or zones (zone control), while some other variables - possibly including input variables - are steered to fixed target or set-point. In real applications, this problem is often overcome by including and excluding an appropriate penalization for the output errors in the control cost function. In this way, throughout the continuous operation of the process, the control system keeps switching from one controller to another, and even if a stabilizing control law is developed for each of the control configurations, switching among stable controllers not necessarily produces a stable closed loop system. From a theoretical point of view, the control objective of this kind of problem can be seen as a target set (in the output space) instead of a target point, since inside the zones there are no preferences between one point or another. In this work, a stable MPC formulation for constrained linear systems, with several practical properties is developed for this scenario. The concept of distance from a point to a set is exploited to propose an additional cost term, which ensures both, recursive feasibility and local optimality. The performance of the proposed strategy is illustrated by simulation of an ill-conditioned distillation column. (C) 2010 Elsevier Ltd. All rights reserved.

Evaluation of FAO Penman-Monteith and alternative methods for estimating reference evapotranspiration with missing data in Southern Ontario, Canada

Grass reference evapotranspiration (ETo) is an important agrometeorological parameter for climatological and hydrological studies, as well as for irrigation planning and management. There are several methods to estimate ETo, but their performance in different environments is diverse, since all of them have some empirical background. The FAO Penman-Monteith (FAD PM) method has been considered as a universal standard to estimate ETo for more than a decade. This method considers many parameters related to the evapotranspiration process: net radiation (Rn), air temperature (7), vapor pressure deficit (Delta e), and wind speed (U); and has presented very good results when compared to data from lysimeters Populated with short grass or alfalfa. In some conditions, the use of the FAO PM method is restricted by the lack of input variables. In these cases, when data are missing, the option is to calculate ETo by the FAD PM method using estimated input variables, as recommended by FAD Irrigation and Drainage Paper 56. Based on that, the objective of this study was to evaluate the performance of the FAO PM method to estimate ETo when Rn, Delta e, and U data are missing, in Southern Ontario, Canada. Other alternative methods were also tested for the region: Priestley-Taylor, Hargreaves, and Thornthwaite. Data from 12 locations across Southern Ontario, Canada, were used to compare ETo estimated by the FAD PM method with a complete data set and with missing data. The alternative ETo equations were also tested and calibrated for each location. When relative humidity (RH) and U data were missing, the FAD PM method was still a very good option for estimating ETo for Southern Ontario, with RMSE smaller than 0.53 mm day(-1). For these cases, U data were replaced by the normal values for the region and Delta e was estimated from temperature data. The Priestley-Taylor method was also a good option for estimating ETo when U and Delta e data were missing, mainly when calibrated locally (RMSE = 0.40 mm day(-1)). When Rn was missing, the FAD PM method was not good enough for estimating ETo, with RMSE increasing to 0.79 mm day(-1). When only T data were available, adjusted Hargreaves and modified Thornthwaite methods were better options to estimate ETo than the FAO) PM method, since RMSEs from these methods, respectively 0.79 and 0.83 mm day(-1), were significantly smaller than that obtained by FAO PM (RMSE = 1.12 mm day(-1). (C) 2009 Elsevier B.V. All rights reserved.

A stochastic frontier production function with flexible risk properties

This paper considers a stochastic frontier production function which has additive, heteroscedastic error structure. The model allows for negative or positive marginal production risks of inputs, as originally proposed by Just and Pope (1978). The technical efficiencies of individual firms in the sample are a function of the levels of the input variables in the stochastic frontier, in addition to the technical inefficiency effects. These are two features of the model which are not exhibited by the commonly used stochastic frontiers with multiplicative error structures, An empirical application is presented using cross-sectional data on Ethiopian peasant farmers. The null hypothesis of no technical inefficiencies of production among these farmers is accepted. Further, the flexible risk models do not fit the data on peasant farmers as well as the traditional stochastic frontier model with multiplicative error structure.

A Bayesian approach to fuzzy hypotheses testing for the estimation of optimal age for vaccination against measles

Fuzzy Bayesian tests were performed to evaluate whether the mother`s seroprevalence and children`s seroconversion to measles vaccine could be considered as ""high"" or ""low"". The results of the tests were aggregated into a fuzzy rule-based model structure, which would allow an expert to influence the model results. The linguistic model was developed considering four input variables. As the model output, we obtain the recommended age-specific vaccine coverage. The inputs of the fuzzy rules are fuzzy sets and the outputs are constant functions, performing the simplest Takagi-Sugeno-Kang model. This fuzzy approach is compared to a classical one, where the classical Bayes test was performed. Although the fuzzy and classical performances were similar, the fuzzy approach was more detailed and revealed important differences. In addition to taking into account subjective information in the form of fuzzy hypotheses it can be intuitively grasped by the decision maker. Finally, we show that the Bayesian test of fuzzy hypotheses is an interesting approach from the theoretical point of view, in the sense that it combines two complementary areas of investigation, normally seen as competitive. (C) 2007 IMACS. Published by Elsevier B.V. All rights reserved.

Controle estatístico de processo aplicado no monitoramento da absorção de água por carcaças de frango durante a etapa de pré-resfriamento

A absorção de água por carcaças de frango na etapa de pré-resfriamento da linha abate representa uma característica de qualidade importante relacionada ao rendimento do produto final. Uma forma de manter o padrão de qualidade de um produto é garantir que as etapas do processo sejam estáveis e replicáveis. Ao empregar o Controle Estatístico de Processo (CEP) é possível obter estabilidade e melhorias nos processos, por meio da redução da variabilidade. Neste contexto, o objetivo deste trabalho foi a aplicação de gráficos de controle, análise de correlação, estatística descritiva, testes de hipóteses e regressão linear múltipla na linha de abate de um abatedouro-frigorífico de aves para monitorar a variabilidade da absorção de água pelas carcaças de frango após a etapa de pré-resfriamento. Como resultado, verificou-se que o teor de absorção de água das carcaças de frango apresentou elevada variabilidade, sendo que 10% (8/80) das carcaças apresentaram absorção de água superior ao limite de 8% definido pela legislação brasileira. Do total de 16 variáveis de entrada analisadas, as mais impactantes no teor de absorção de água foram o “tempo de retenção da carcaça no pré-chiller” e o “tempo de espera da carcaça após a etapa de gotejamento”. Entretanto, o modelo de regressão obtido apresentou baixa correlação (R²=0,16) que foi associada à elevada variabilidade da variável-resposta. Os resultados da estatística descritiva demonstraram que as variáveis de entrada também apresentaram elevada variabilidade, com coeficiente de variação entre 7,95 e 63,5%. Verificou-se, pela análise dos gráficos de controle de medida individual e da amplitude móvel, que 15 das 16 variáveis de entrada se apresentaram fora de controle estatístico assim como a variável-resposta. Baseado no fluxograma e na descrição das etapas da linha de abate, previamente realizados, atribuiu-se à falta de padronização na condução das etapas e de procedimentos para o controle de qualidade das operações na linha de abate como fatores relevantes que poderiam estar associados à presença de causas especiais no processo. Concluiu-se que para reduzir a elevada variabilidade das variáveis e eliminar as causas especiais presentes são necessários ajustes operacionais para, dessa forma, obter um processo mais estável e mais uniforme garantindo o padrão de qualidade das carcaças de frango em relação ao teor de absorção de água.