• Environmental Engineering Research
• /
• v.14 no.3
• /
• pp.186-194
• /
• 2009

Understanding the environmental fate of human and animal pharmaceuticals and their risk assessment are of great importance due to their growing environmental concerns. Although there are many potential pathways for them to reach the environment, effluents from sewage treatment plants (STPs) are recognized as major point sources. In this study, the removal efficiencies of the 43 selected priority pharmaceuticals in a conventional STP were evaluated using two simple models: an equilibrium partitioning model (EPM) and STPWIN$^{TM}$ program developed by US EPA. It was expected that many pharmaceuticals are not likely to be removed by conventional activated sludge processes because of their relatively low sorption potential to suspended sludge and low biodegradability. Only a few pharmaceuticals were predicted to be easily removed by sorption or biodegradation, and hence a conventional STP may not protect the environment from the release of unwanted pharmaceuticals. However, the prediction made in this study strongly relies on sorption coefficient to suspended sludge and biodegradation half-lives, which may vary significantly depending on models. Removal efficiencies predicted using the EPM were typically higher than those predicted by STPWIN for many hydrophilic pharmaceuticals due to the difference in prediction method for sorption coefficients. Comparison with experimental organic carbon-water partition coefficients ($K_{ocs}) revealed that log KOW-based estimation used in STPWIN is likely to underestimate sorption coefficients, thus resulting low removal efficiency by sorption. Predicted values by the EPM were consistent with limited experimental data although this model does not include biodegradation processes, implying that this simple model can be very useful with reliable Koc values. Because there are not many experimental data available for priority pharmaceuticals to evaluate the model performance, it should be important to obtain reliable experimental data including sorption coefficients and biodegradation rate constants for the prediction of the fate of the selected pharmaceuticals.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.17 no.1
• /
• pp.7-13
• /
• 2011

The behavior prediction of PAHs in Jinhae Bay using a three-dimensional ecological model(EMT-3D) was examined. A three-dimensional ecological model(EMT-3D) was applied to the simulation of PAHs behaviors in Jinhae Bay of Korea. The computed results of simulation were in good agreement with the observed values. The result of sensitivity analysis showed that photolysis coefficient and extinction coefficient were important factors in the variation of dissolved PAHs, and POC partition coefficient was important factor in the variation of PAHs in particulate organic matter. In the case of PAHs in phytoplankton, bioconcentration factor of plankton was the most significant and the most effective in all. In simulations of 30%, 50% and 80% reduction in total loads of PAHs, the concentrations of dissolved PAHs were shown to be lower than 24 ng/L, 20 ng/L and 16 ng/L, respectively.

• Earthquakes and Structures
• /
• v.6 no.2
• /
• pp.141-161
• /
• 2014

This article presents a new generation of empirical ground motion models for the prediction of response spectral accelerations in soil conditions, specifically developed for the Vrancea intermediate-depth seismic source. The strong ground motion database from which the ground motion prediction model is derived consists of over 800 horizontal components of acceleration recorded from nine Vrancea intermediate-depth seismic events as well as from other seventeen intermediate-depth earthquakes produced in other seismically active regions in the world. Among the main features of the new ground motion model are the prediction of spectral ordinates values (besides the prediction of the peak ground acceleration), the extension of the magnitudes range applicability, the use of consistent metrics (epicentral distance) for this type of seismic source, the extension of the distance range applicability to 300 km, the partition of total standard deviation in intra- and inter-event standard deviations and the use of a national strong ground motion database more than two times larger than in the previous studies. The results suggest that this model is an improvement of the previous generation of ground motion prediction models and can be properly employed in the analysis of the seismic hazard of Romania.

• Journal of Soil and Groundwater Environment
• /
• v.9 no.1
• /
• pp.28-38
• /
• 2004

This study presents the result of uncertainty and sensitivity analysis of a pharmacokinetic model which describes the distribution and removal of benzene at each organ when an indivisual inhales indoor contaminated air with benzene originated from groundwater. The pharmacokinetic model simulates the distribution of benzene deposited in organs of human body through inhalation of contaminated indoor air as well as degradation-metabolism in liver. This study focused on the uncertainty problem induced from the use of the single values for blood flow, partition coefficient, degradation constant, volume, etc. of each organ which was due to a lack of knowledge about these parameters or their measurements. To solve this problem, uncertainty analysis on the pharmacokinetic model was conducted simultaneously which would help understanding the risk assessment associated with VOCs.

• Journal of Soil and Groundwater Environment
• /
• v.8 no.2
• /
• pp.44-54
• /
• 2003

A kinetic model for evaluating effects of surfactant on the biodegradation of HOC(hazardous organic chemicals) in soil-slurry systems was developed. The model includes the partition of HOC and surfactant, the dissolved-, micellar-, and sorbed-phase biodegradation, the enhanced solubilization of HOC by surfactant addition, and the mass transfer of HOC. Phenanthrene as HOC and Trition X-100, Tergitol NP-10, Igepal CA-720, and Brij 30 were used in the model simulations. The biodegradation rate was increased even with a small micellera-phase bioavailability. The biodegradation was not greatly enhanced due to decreased aqueous HOC concentration by increasing surfactant dose in both cases with and without micellar-phase bioavailability. The effect of sorbed-phase biodegradation on total biodegradation rate was not highly important compared to aqueous- and micellar-phase biodegradation. The model can be applied for surfactant screening and optimal design of surfactant-based soil bioremediation process.

• The Korean Journal of Applied Statistics
• /
• v.30 no.1
• /
• pp.57-68
• /
• 2017

Clustering algorithms attempt to find a partition of a finite set of objects in to a potentially predetermined number of nonempty subsets. Gibbs sampling of a normal mixture of linear mixed regressions with a Dirichlet prior distribution calculates posterior probabilities when the number of clusters was known. Our approach provides simultaneous partitioning and parameter estimation with the computation of classification probabilities. A Monte Carlo study of curve estimation results showed that the model was useful for function estimation. Examples are given to show how these models perform on real data.

• The Korean Journal of Applied Statistics
• /
• v.34 no.2
• /
• pp.153-166
• /
• 2021

A main goal of pharmacogenomics studies is to predict individual's drug responsiveness based on high dimensional genetic variables. Due to a large number of variables, feature selection is required in order to reduce the number of variables. The selected features are used to construct a predictive model using machine learning algorithms. In the present study, we applied several hybrid feature selection methods such as combinations of logistic regression, ReliefF, TurF, random forest, and LASSO to a next generation sequencing data set of 400 epilepsy patients. We then applied the selected features to machine learning methods including random forest, gradient boosting, and support vector machine as well as a stacking ensemble method. Our results showed that the stacking model with a hybrid feature selection of random forest and ReliefF performs better than with other combinations of approaches. Based on a 5-fold cross validation partition, the mean test accuracy value of the best model was 0.727 and the mean test AUC value of the best model was 0.761. It also appeared that the stacking models outperform than single machine learning predictive models when using the same selected features.

• Steel and Composite Structures
• /
• v.44 no.1
• /
• pp.33-47
• /
• 2022

This paper presents a finite element model for predicting the monotonic behaviour of bolted endplate joints connecting steel-concrete composite walls and steel beams. The demountable Hollo-bolts are utilised to facilitate the quick installation and dismantling for replacement and reuse. In the developed model, material and geometric nonlinearities were included. The accuracy of the developed model was assessed by comparing the numerical results with previous experimental tests on hollow/composite column-to-steel beam joints that incorporated endplates and Hollo-bolts. In particular, the Hollo-bolts were modelled with the expanded sleeves involved, and different material properties of the Hollo-bolt shank and sleeves were considered based on the information provided by the manufacture. The developed models, therefore, can be applied in the present study to simulate the wall-to-beam joints with similar structural components and characteristics. Based on the validated model, the authors herein compared the behaviour of wall-to-beam joints of two commonly utilised composite walling systems (Case 1: flat steel plates with headed studs; Case 2: lipped channel section with partition plates). Considering the ease of manufacturing, onsite erection and the pertinent costs, composite walling system with flat steel plates and conventional headed studs (Case 1) was the focus of present study. Specifically, additional headed studs were pre-welded inside the front wall plates to enhance the joint performance. On this basis, a series of parametric studies were conducted to assess the influences of five design parameters on the behaviour of bolted endplate wall-to-beam joints. The initial stiffness, plastic moment capacity, as well as the rotational capacity of the composite wall-to-beam joints based on the numerical analysis were further compared with the current design provision.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.4B
• /
• pp.375-381
• /
• 2008

In this research, a physics based grid-multi layer distributed flood runoff model was developed to analyze discharge for the Namgang Dam Watershed ($2,293km^2$) and applied for sensitivity analysis for estimation of parameters, mainly initial soil moisture condition and saturate infiltration coefficient, which have a strong influence on discharge. Capability of the model was evaluated using VER and QER from the results of rainfall-runoff analysis and showed enhanced results of 6% compared to parameters before calibration. As the result with the sensitivity analysis of parameters, the part of the most influence on the runoff was the infiltration coefficient and ratio of layer partition. The total discharge and peak time showed comparatively precise runoff results without the initial calibration of the parameters.

• Korean Chemical Engineering Research
• /
• v.61 no.4
• /
• pp.523-541
• /
• 2023

As accessibility to 3D printers increases, there is a growing frequency of exposure to chemicals associated with 3D printing. However, research on the toxicity and harmfulness of chemicals generated by 3D printing is insufficient, and the performance of toxicity prediction using in silico techniques is limited due to missing molecular structure data. In this study, quantitative structure-activity relationship (QSAR) model based on data-centric AI approach was developed to predict the toxicity of new 3D printing materials by imputing missing values in molecular descriptors. First, MissForest algorithm was utilized to impute missing values in molecular descriptors of hazardous 3D printing materials. Then, based on four different machine learning models (decision tree, random forest, XGBoost, SVM), a machine learning (ML)-based QSAR model was developed to predict the bioconcentration factor (Log BCF), octanol-air partition coefficient (Log Koa), and partition coefficient (Log P). Furthermore, the reliability of the data-centric QSAR model was validated through the Tree-SHAP (SHapley Additive exPlanations) method, which is one of explainable artificial intelligence (XAI) techniques. The proposed imputation method based on the MissForest enlarged approximately 2.5 times more molecular structure data compared to the existing data. Based on the imputed dataset of molecular descriptor, the developed data-centric QSAR model achieved approximately 73%, 76% and 92% of prediction performance for Log BCF, Log Koa, and Log P, respectively. Lastly, Tree-SHAP analysis demonstrated that the data-centric-based QSAR model achieved high prediction performance for toxicity information by identifying key molecular descriptors highly correlated with toxicity indices. Therefore, the proposed QSAR model based on the data-centric XAI approach can be extended to predict the toxicity of potential pollutants in emerging printing chemicals, chemical process, semiconductor or display process.