GBGermán Bianchini

Germán Bianchini

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Germán Bianchini has a total of 38 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Evolutionary Statistical System Based on Novelty Search: A Parallel Metaheuristic for Uncertainty Reduction Applied to Wildfire Spread Prediction

    by Jan Strappa, Paola Caymes-Scutari, Germán Bianchini

    The problem of wildfire spread prediction presents a high degree of complexity due in large part to the limitations for providing accurate input parameters in real time (e.g., wind speed, temperature, moisture of the soil, etc.). This uncertainty in the environmental values has led to the development of computational methods that search the space of possible combinations of parameters (also called scenarios) in order to obtain better predictions. State-of-the-art methods are based on parallel optimization strategies that use a fitness function to guide this search. Moreover, the resulting predictions are based on a combination of multiple solutions from the space of scenarios. These methods have improved the quality of classical predictions; however, they have some limitations, such as premature convergence. In this work, we evaluate a new proposal for the optimization of scenarios that follows the Novelty Search paradigm. Novelty-based algorithms replace the objective function by a measure of the novelty of the solutions, which allows the search to generate solutions that are novel (in their behavior space) with respect to previously evaluated solutions. This approach avoids local optima and maximizes exploration. Our method, Evolutionary Statistical System based on Novelty Search (ESS-NS), outperforms the quality obtained by its competitors in our experiments. Execution times are faster than other methods for almost all cases. Lastly, several lines of future work are provided in order to significantly improve these results.


  • Open Access

    A Parallel Novelty Search Metaheuristic Applied to a Wildfire Prediction System

    by Jan Strappa, Paola Caymes-Scutari, Germán Bianchini

    Wildfires are a highly prevalent multi-causal environmental phenomenon. The impact of this phenomenon includes human losses, environmental damage and high economic costs. To mitigate these effects, several computer simulation systems have been developed in order to predict fire behavior based on a set of input parameters, also called a scenario (wind speed and direction; temperature; etc.). However, the results of a simulation usually have a high degree of error due to the uncertainty in the values of some variables, because they are not known, or because their measurement may be imprecise, erroneous, or impossible to perform in real time. Previous works have proposed the combination of multiple results in order to reduce this uncertainty. State-of-the-art methods are based on parallel optimization strategies that use a fitness function to guide the search among all possible scenarios. Although these methods have shown improvements in the quality of predictions, they have some limitations related to the algorithms used for the selection of scenarios. To overcome these limitations, in this work we propose to apply the Novelty Search paradigm, which replaces the objective function by a measure of the novelty of the solutions found, which allows the search to continuously generate solutions with behaviors that differ from one another. This approach avoids local optima and may be able to find useful solutions that would be difficult or impossible to find by other algorithms. As with existing methods, this proposal may also be adapted to other propagation models (floods, avalanches or landslides).

  • A Parallel Novelty Search Metaheuristic Applied to a Wildfire Prediction System

    by Jan Strappa, Paola Caymes-Scutari, German Bianchini

    Wildfires are a highly prevalent multi-causal environmental phenomenon. The impact of this phenomenon includes human losses, environmental damage and high economic costs. To mitigate these effects, several computer simulation systems have been developed in order to predict fire behavior based on a set of input parameters, also called a scenario (wind speed and direction; temperature; etc.). However, the results of a simulation usually have a high degree of error due to the uncertainty in the values of some variables, because they are not known, or because their measurement may be imprecise, erroneous, or impossible to perform in real time. Previous works have proposed the combination of multiple results in order to reduce this uncertainty. State-of-the-art methods are based on parallel optimization strategies that use a fitness function to guide the search among all possible scenarios. Although these methods have shown improvements in the quality of predictions, they have some limitations related to the algorithms used for the selection of scenarios. To overcome these limitations, in this work we propose to apply the Novelty Search paradigm, which replaces the objective function by a measure of the novelty of the solutions found, which allows the search to continuously generate solutions with behaviors that differ from one another. This approach avoids local optima and may be able to find useful solutions that would be difficult or impossible to find by other algorithms. As with existing methods, this proposal may also be adapted to other propagation models (floods, avalanches or landslides).


  • Open Access

    Método de Reducción de Incertidumbre Basado en Algoritmos Evolutivos y Paralelismo Orientado a la Predicción y Prevención de Desastres Naturales

    by Miguel Méndez Garabetti, Germán Bianchini, Verónica Gil Costa, Paola Caymes Scutari

    La falta de precisión en los parámetros de entrada en cualquier modelo científico suele ser habitual debido a la imposibilidad de cuantificar todos los factores que afectan al modelo en tiempo real. Esto puede producir consecuencias dramáticas en la salida del mismo si se trata de un sistema crítico, tal como lo es en el caso particular de la predicción de fenómenos naturales. No obstante, también deben sumarse las limitaciones impuestas por los propios modelos, las restricciones que agregan las soluciones numéricas y las provenientes de las propias implementaciones y versiones informáticas. Por tal motivo, esta investigación se ha centrado en el desarrollo de métodos informáticos que se enfoquen en el tratamiento de la incertidumbre de dichos valores de entrada para lograr así una predicción lo más confiable posible por parte del modelo. En base a esto, se han desarrollado dos métodos de reducción de incertidumbre que han sido validados en la predicción del comportamiento de incendios forestales. Dichos métodos utilizan como componentes: Análisis Estadístico, Algoritmos Evolutivos Paralelos, Computación de Alto Desempeño e Información Satelital. El análisis estadístico ofrece la posibilidad de estudiar tendencias en el comportamiento del sistema bajo diversas condiciones, los algoritmos evolutivos paralelos permiten orientar la búsqueda brindando una predicción más precisa, la computación paralela/distribuida brinda infraestructura necesaria para poder llevar a cabo el procesamiento en cantidad de tiempo razonable y la información satelital permite aplicar estos métodos situaciones de emergencia real. Cada uno de estos componentes son indispensables para lograr valores aceptables tanto en términos de calidad de predicción como términos de tiempo de respuesta de los métodos propuestos.


  • Dynamic Tuning of a Forest Fire Prediction Parallel Method

    by Paola Caymes-Scutari, María Laura Tardivo, Germán Bianchini, Miguel Méndez-Garabetti


  • Optimization for an Uncertainty Reduction Method Applied to Forest Fires Spread Prediction

    by María Laura Tardivo, Paola Caymes-Scutari, Miguel Méndez-Garabetti, Germán Bianchini


  • Hierarchical parallel model for improving performance on differential evolution

    by María Laura Tardivo, Paola Caymes‐Scutari, Germán Bianchini, Miguel Méndez‐Garabetti

    Summary: This paper presents a parallel distributed model for the Differential Evolution algorithm. The proposed model, Hierarchical Island‐Based Model for Differential Evolution, follows a double‐hierarchy master‐worker scheme and offers two parallelism levels. In this proposal, the processes are associated with certain cooperation hierarchy, allowing them to explore in a more comprehensive way the search space of the problem at hand. A comparative study with other algorithms from the state of art is also presented. The results show that Hierachical Island‐Based Model for Differential Evolution is a flexible model and achieves good performance in terms of results quality and computing time.


  • ESS-IM applied to forest fire spread prediction: Parameters tuning for a heterogeneous configuration

    by Miguel Mendez-Garabetti, German Bianchini, Paola Caymes-Scutari, Maria Laura Tardivo

    Forest fires are a critical natural hazard in many regions of the World. For this reason, the prediction of this kind of phenomenon is considered a very important task that involves a high degree of complexity and precision. The ability to predict the forest fire behaviour constitutes an important tool for managers, helping to improve the effectiveness of fire prevention, detection and firefighting resources allocation. For this reason, prediction methods should be configured to operate as efficiently as possible. In this paper, a calibration study of Evolutionary-Statistical System with Island Model's evolutionary parameters is presented (ESS-IM). ESS-IM is a general-parallel uncertainty reduction method applied to the forest fires spread prediction.


  • Open Access

    A comparative study of evolutionary statistical methods for uncertainty reduction in forest fire propagation prediction

    by María Laura Tardivo, Paola Caymes-Scutari, Germán Bianchini, Miguel Méndez-Garabetti, Andrés Cencerrado, Ana Cortés