Component importance measures are relevant to improve the system design and to develop optimal replacement policies. Birnbaum's importance measure is one of the most relevant measures. If the ...components are (stochastically) independent, this measure can be defined using several equivalent expressions. However, in many practical situations, the independence assumption is unrealistic. It also turns out that in the case of dependent components, different Birnbaum's measure definitions lead to different concepts. In this paper, we extend Birnbaum's importance measure to the case of dependent components in a way allowing us to obtain relevant properties including connections and comparisons with other measures proposed and studied recently. The dependence is modeled through copulas and the new measure is based on the contribution of the component to the system reliability.
•The paper presents an active-learning oriented importance sampling method.•Importance sampling centers are optimized by using the active learning mechanism.•The surrogate model can cover all ...branches of the system limit-state surface.•Several case studies have demonstrated engineering applications of the approach.
A major issue in the structural reliability analysis is to determine an accurate estimation result of the failure probability ideally based on a small number of model evaluations. In this regard, the active-learning Kriging based importance sampling method has been received considerable attentions. However, the utility of the most probable failure point (MPP) as the unique sampling center has limited its potential applications for multi-MPP problems. To this end, the paper presents an adaptive Kriging oriented importance sampling (AKOIS) approach. The outer active-learning loop of the AKOIS procedure is used to identify importance sampling centers, whereas its inner-loop is realized based on a rather small subregion centering at the sampling center to gain new training samples. Besides, numerical convergence of local active-learning iterations will immediately trigger another round of outer global search for a new importance sampling center. In this regard, the determined importance sampling centers are able to adaptively cover all branches of the investigated limit-state surface for structural system reliability analysis. Engineering applications of the data-driven importance sampling approach are demonstrated by several system reliability examples in the literature.
Though reported capture fisheries are dominated by marine production, inland fish and fisheries make substantial contributions to meeting the challenges faced by individuals, society, and the ...environment in a changing global landscape. Inland capture fisheries and aquaculture contribute over 40% to the world's reported finfish production from less than 0.01% of the total volume of water on earth. These fisheries provide food for billions and livelihoods for millions of people worldwide. Herein, using supporting evidence from the literature, we review 10 reasons why inland fish and fisheries are important to the individual (food security, economic security, empowerment), to society (cultural services, recreational services, human health and well-being, knowledge transfer and capacity building), and to the environment (ecosystem function and biodiversity, as aquatic “canaries”, the “green food” movement). However, the current limitations to valuing the services provided by inland fish and fisheries make comparison with other water resource users extremely difficult. This list can serve to demonstrate the importance of inland fish and fisheries, a necessary first step to better incorporating them into agriculture, land-use, and water resource planning, where they are currently often underappreciated or ignored.
To ensure safe drinking water sources in the future, it is imperative to understand the quality and pollution level of existing groundwater. The prediction of water quality with high accuracy is the ...key to control water pollution and the improvement of water management. In this study, a deep learning (DL) based model is proposed for predicting groundwater quality and compared with three other machine learning (ML) models, namely, random forest (RF), eXtreme gradient boosting (XGBoost), and artificial neural network (ANN). A total of 226 groundwater samples are collected from an agriculturally intensive area Arang of Raipur district, Chhattisgarh, India, and various physicochemical parameters are measured to compute entropy weight-based groundwater quality index (EWQI). Prediction performances of models are determined by introducing five error metrics. Results showed that DL model is the best prediction model with the highest accuracy in terms of R2, i.e., R2 = 0996 against the RF (R2 = 0.886), XGBoost (R2 = 0.0.927), and ANN (R2 = 0.917). The uncertainty of the DL model output is cross-verified by running the proposed algorithm with newly randomized dataset for ten times, where minor deviations in the mean value of performance metrics are observed. Moreover, input variable importance computed by prediction models highlights that DL model is the most realistic and accurate approach in the prediction of groundwater quality.
•Groundwater quality is assessed using EWQI method.•Machine learning (ML) algorithms are used for predicting groundwater quality.•Prediction performance of RF, XGBoost, ANN and DL models are compared.•DL based quality prediction model performs much better than other ML models.
•A new methodology to evaluate the stability of WBFS is proposed.•A voting approach was used to compare different importance rankings.•Experiments on simulated and real-world datasets.•Existing WBFS ...methods are not stable on real-world datasets.
Weight-based feature selection (WBFS) are methods used to measure the contribution of input to output in a trained artificial neural network (ANN). Furthermore, algorithms such as Garson’s rely upon a single best neural network model or the mean importance value of several ANNs. However, different initialization weights lead to different importance values, as reported in other studies. These differences are misleading since each rank could result in different scores, altering the position of a variable in a given rank. Therefore, we propose a new methodology to assess the stability of a WBFS method. In essence, the idea is to use a voting approach to evaluate the importance of rankings. The results showed that Garson’s, Olden’s and Yoon’s algorithms are more stable methods when applied to artificial datasets. Nevertheless, its stability is considerably reduced when applied to real-world datasets. Hence, we concluded that future work should take into consideration the aforementioned instability of existing WBFS methods as applied to complex real-world data.
In this paper, the adaptive importance sampling (AIS) method is extended for hybrid reliability analysis under random and interval variables (HRA-RI) with small failure probabilities. In AIS, the ...design space is divided into random and interval variable subspaces. In random variable subspace, Markov Chain Monte Carlo (MCMC) is employed to generate samples which populate the failure regions. Then based on these samples, two kernel sampling density functions are established for estimations of the lower and upper bounds of failure probability. To improve the computational efficiency of AIS in cases with time-consuming performance functions, a combination method of projection-outline-based active learning Kriging and AIS, termed as POALK-AIS, is proposed in this paper. In this method, design of experiments is sequentially updated for the construction of Kriging metamodel with focus on the approximation accuracy of the projection outlines on the limit-state surface. During the procedure of POALK-AIS, multiple groups of sample points simulated by AIS are used to calculate the upper and lower bounds of failure probability. The accuracy, efficiency and robustness of POALK-AIS for HRA-RI with small failure probabilities are verified by five test examples.
•AIS method is extended for HRA-RI with small failure probabilities.•A combination method of POALK and AIS is proposed.•Approximation accuracy of projection outlines is improved by refining Kriging model.•The performance of POALK-AIS for HRA-RI with small failure probabilities is verified.
We report for the first time conical galls of Clinodiplosis capsici Gagné, 2000 (Diptera, Cecidomyiidae) on sweet pepper (Capsicum annuum L., Solanaceae) in Brazil. This report has ...agricultural importance since this midge is one of the known sweet-pepper pests.
The factors influencing residents health have become complex and intertwined with the development of economy and society. Traditional research with a single factor on health will not provide an ...accurate picture of the situation. This paper collects data on economic, environmental and social factors to estimate their impact on regional health. Considering the data is multi-source and complex, this paper proposes a combined feature importance algorithm, which weighted the feature importance of RF, XGB and SOIL. The algorithm does not depend on the data and adaptively approximates the true results. The results show that economic factors have a significant and direct impact on health, environmental factors have a lag correlation with health level, and social factors have a more complicated effect on health. Finally, we provide policy suggestions for health on economic, environmental, and social development.