China
Africa
Explore chapters and articles related to this topic
Agent-Based Models for Water-Related Disaster Risk Management
Published in Neiler de Jesús Medina Peña, Adaptive Disaster Risk Assessment Combining Multi-Hazards With Socioeconomic Vulnerability and Dynamic Exposure, 2021
Most of the SA reported in our review used the OFAT (One-Factor-At-A-Time) method. We believe it is because the method is relatively easy to perform and yet give valuable insights. For example, Tonn and Guikema (2018 – Supplemental online material), assessed the sensitivity of seven input parameters using the OFAT approach. Similarly, Yang et al. (2018) present a SA on six strategic Variables for flood loss assessment. Also, Baeza et al. (2019) use SA to evaluate the degree of sensitivity of governance scenarios in the model outputs for different regions in Mexico City. Only two papers from our review performed global sensitivity analysis (GSA): in Erdlenbruch and Bonté (2018) applied GSA to evaluate the dynamics of individual adaptation to floods, and Dressler et al. (2016) applied GSA to assess the robustness of sub-models over an extensive parameter range.
Global sensitivity analysis in the design of rockfill dams
Published in Jean-Pierre Tournier, Tony Bennett, Johanne Bibeau, Sustainable and Safe Dams Around the World, 2019
Global SA can broadly be classified into model based and model free methods. Examples of model-based techniques include One-Factor-At-a-Time Method (OAT) and those that use regression. Some of their main shortcomings are incorrect sensitivity estimates due to a regression equation that does not fit well with the response surface, and the inability to measure and detect interactions among different variables. Regional sensitivity analysis and variance based methods are important examples of model free techniques, with the latter being the most advanced and intuitive approach (Guo et al 2018; Lilburne et al 2009). The global SA considers the entire range of uncertainty associated with the inputs as characterized by their joint probability density functions. In a global SA, all the inputs are varied simultaneously (not in OAT) and usually they are considered independent. This method requires more computational work when compared to the local approach; however, the former is better suited to deal with nonlinear responses. The fundamental steps that constitute the global SA technique are: i) specification of the computational model, ii) determination of relevant inputs and their bounds, iii) input sample generation by a sampling design method, iv) model evaluation with the generated input parameters; and v) uncertainty analysis and calculation of relative importance of each input through a sensitivity estimator.
Optimization of media composition for enhancing tetracycline degradation by Trichosporon mycotoxinivorans XPY-10 using response surface methodology
Published in Environmental Technology, 2021
Xiaochen Huang, Xinyang Zhang, Yanyan Huang, Xuping Xu
In our previous study, variables (carbon source, nitrogen source, inorganic salt and pH value) strongly affected the ability of XPY-10 to degrade tetracycline and growth of cells were studied using traditional single factor test method [12]. However, this conventional one-factor-at-a-time method could not explain the interactions of different variables and achieve the optimum conditions of variables [13]. In order to achieve high tetracycline biodegradation efficiency, it is necessary to apply a multivariate optimization strategy for the degradation conditions. Statistical optimization method such as response surface methodology (RSM) and contour plot method have been applied for optimization some bioprocesses [14, 15]. RSM provides a fast and effective way to identify the important factors and select the optimum experimental conditions of variables than the classical single-factor test [16]. However, limited report is available regarding tetracycline biodegradation.
Textile applications of commercial photochromic dyes: part8. A statistical investigation of the influence of photochromic dyes on thermoplastic fibres using a UV-irradiation technique
Published in The Journal of The Textile Institute, 2020
Basel Younes, Stephanie C. Ward, Robert M. Christie
Statistical experimental design analysis solves the problems that arise in traditional analyses depending on the one factor-at-a-time method. The process factors are classified as controllable, noise and constant factors. The noise factors are classified thus (Phadke, 1989): first, external factors which are environmental noise factors like temperature, humidity, dust, supply voltage, electromagnetic interaction, vibrations of instrument supports or in the sunlight intensity through the windows and human errors; secondly, unit-to-unit variations like resistance; finally, deterioration in the product as time passes. The focus of SED is the optimizing of the average response values depending on the factors and their levels. Design and analysis of experimental methods add additional advantage to the analysis of manufacturing processes such as a fibre-to-fabric engineering‘approach (El-Mogahzy, 2009).
Optimization of the freeze-drying process for microemulsion systems
Published in Drying Technology, 2019
Andreza Rochelle do Vale Morais Morais, Francisco Humberto Xavier-Jr., Éverton do Nascimento Alencar, Christian Melo de Oliveira, Nednaldo Dantas Santos, Arnóbio Antônio Silva-Júnior, Gillian Barratt, Eryvaldo Sócrates Tabosa do Egito
DOE is a statistical approach used to determine the influence of several independent variables on the dependent variable of the process. The optimal design allows the time and cost of the experimentation to be reduced, as well as improving the process yield.[15] Therefore, this method is used much more often than the one factor at a time method, which is time-consuming and expensive because it requires a large number of experiments and does not examine interactions between the variables.[16] The response surface methodology is a technique of DOE that combines mathematics and statistics to analyze the relative significance of different parameters, finding the optimal working conditions, by combining a small number of variables, resulting in fewer experiments.[17]