Background and Objective: Ground level ozone (O3) is one of most dangerous pollutants for human health in urban areas. The aim of this study was to identify the factors affecting the formation of ozone and modeling the spatial and temporal variations of ozone concentration in Tehran metropolitan area.Materials and Methods: The data used in this research included meteorological data and pollution concentration data for 2014. First, we studied the impact and correlation of parameters to ozone concentration using the coefficient of Pearson, and then we did modeling of ozone concentration using a Multivariate Linear Regression method.Results: The developed model had the ability to describe 79% of the data changes for 2014. The temporal analysis of the ozone concentration showed that the best coefficient of determination of the model was R2=0.771 in the summer and R2=0.778 in July. These results also showed that among the air quality monitoring station of Tehran, station 4 had the lowest coefficient of determination (R2=0.6) and Aqdasieh station had the highest coefficient of determination (R2=0.79). Finally, the spatial distribution of the estimated ozone concentration was consistent with the measured ozone concentration at the station level.Conclusion: According to the results, all the parameters related to air pollution concentration and meteorological parameters were effective parameters on modeling of ozone concentration on the ground level. The spatial distribution of ozone concentration in Tehran showed a greater concentration of ozone in the South and East than the North and West of the city.

Accurate travel time prediction is one of the important issues in the field of traffic and transportation that can significantly affect the daily life of people and organizations. In this research, four different machine learning methods including Linear Regression, Multivariate Regression, random forest and deep artificial neural network were trained to predict travel time. The purpose of this research is to predict travel time for use in intelligent traffic systems and to use and compare several new methods, including deep neural network and random forest Regression, as well as considering new parameters in the computations such as weather conditions, traffic flow, travel time, and accidents and the traffic locking points compared to other studies are the innovation and comprehensiveness of this study compared to other studies. In the design and implementation of this research, real traffic data taken from Google map was used and analyzed. This data includes information such as traffic conditions, season, time of day, weather conditions, and route characteristics. The results of this research show that the deep neural network (DNN) model with R2 equal to 0.833 has a very good performance among the investigated models. This model explains 0.833% of the variance of the data and the distribution of the residuals in it is relatively central with a mean of zero and a distribution close to normal. The Linear Regression model with R2 equal to 0.615 has a poorer performance than DNN and explains 0.615% of the data variance. But the random Regression model with R2 equal to 0.955 has one of the best performances in competition with DNN and explains 0.955% of the data variance. MSE and RMSE parameters were also used to evaluate the performance of the models, and as a result, a multidimensional comparison was made between the models, and the random forest model resulted in the lowest error values. Since in the collected traffic data, traffic accidents and consequently traffic locking points are also used in the models, and considering that the random forest model is more effectively adapted to the data despite the presence of noise and anomaly, the R2 value of this model is higher than R2 of Deep neural networks, due to the overfitting nature of Deep Learning methods.

We develop a new empirical Bayes analysis in multiple Regression models. In the present work we consider Multivariate skew normal as prior for coefficients of the model in a skew-normal population and give empirical Bayes estimation for parameters of the model. The marginal distribution of response is found to be a closed skew-normal distribution. The empirical Bayes estimator is found in a closed form and the model is applied on a data set.

Summary Knowing geomechanical parameters of soil and rock is among the important items of designing and constructing engineering structures. Deformation modulus may be determined through in-situ tests and indirect methods. For indirect estimating of this module, empirical relations are simple and inexpensive methods, but their uses in other parts of the world were associated with errors due to the variation of rock type and the nature of rock mass. In the present article we attempted to estimate the deformation modulus (Em) of the rock masses of southwestern Iran by using intact rock elastic modulus (Ei) and Rock Mass Rating (RMR) parameters. To do that, the multivariable Linear Regression method was used. The database used included 333 data. In order to study the relation performance and to evaluate its accuracy, R2 coefficient (coefficient of determination) and RMSE (root-mean-square error) were used. For this data R2 coefficient was 0. 811 and RMSE value was 0. 1921...

Creating of resistant and anti-corrosion coatings in nano dimensions are widely used in various industries. The quality of the coating is related to the collision of the nano droplet on the surface and then spreading on it. In many cases, the oblique surface is in the front of the spray nozzle, and then the nano droplet collides obliquely with a surface. Due to expensive and time consuming of the experiments and simulations, model determination for illustrating the effects of the factors on the nano-droplet spreading is very important. In this research, the Multivariate Regression model is being proposed for predicting the nano-droplet spreading data. The nano-droplet spreading has been depended to the speed and impact angle on the surface and that for this reason five models have been considered. The results for comparing the provided models show that the proposed non-Linear Regression has the most efficient and lowest error and has a high fit with optimal output. Also, the residual analysis of the proposed model accepts the normality assumption. Moreover, the correlation between the speed and nano-droplet spreading is 0. 95 which is at a very high level.

Due to the role of recovery in calculating the economic value of ore blocks and the impact of the block's economic value on the design calculations of the final pit and production planning, determination of the amount of metal recovery from the ore material sent to the processing plant is very important. The aim of this study is to investigate the capability of estimating the recovery rate of ore in qualitative manner with three methods based on data classification from data mining techniques and quantitatively using Multivariate Regression and artificial neural networks. Hence, the Miduk copper mine was studied using 58 analyzed samples of the feed of the plant, including Cu, CuO and CuS grades, and the recovery rate of Cu in the final product of the plant. The process of predicting the total recovery of the reserve was made qualitatively by decision tree method, classification based on Bayes rule and k-nearest neighbor (kNN) classification algorithm. For quantitative estimation of recovery, Multivariate Regression and artificial neural network models were established between the mentioned grade parameters and recovery rates (For 47 samples of 58 samples) and with the 11 additional analyzed samples, the obtained models were validated. The coefficient of (R2), Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) in the Regression model were 0. 77, 0. 027722 and 0. 029722, respectively, and in the artificial neural network model, 0. 82, 0. 015753 and 0. 024040, respectively. Therefore, the artificial neural networks model acts as a more accurate tool for predicting recovery versus the multivariable Regression model. The results of sensitivity analysis of artificial neural network model showed that Cu grade is the most important factor and grade of CuO and CuS, respectively, as well as other factors influencing the changes in recovery rate.

Infiltration is the most important soil physical characteristic which its direct measurement is laborious, time consuming and expensive. The purpose of this study is to estimate steady infilterability rate, using Neuro-Fuzzy, Neural Network and Multivariate Linear Regression models. Consequently, steady infilterability rate, was measured using double rings in 100 points in Dehgolan region, Kurdistan Province, Iran. Soil physical (porosity, bulk density, sand, silt and clay) and topography characteristics were measured as readily available properties and used to estimate steady infilterability rate, The data were divided into two sets of terrain (70% of the data) and test (30% of the data). The models based on input type were categorized into type 1 (physical characteristics) and 2 (soil physical and topography characteristics). The results based on mean bias error (MBE), Root Mean Square Error (RMSE), Mean Error (ME), Relative Standard Error (RSE) and Relative Improvement (RI) showed that the Neuro-Fuzzy model (type 1 respectively with statistics 0. 24, 2. 01, 0. 46, 4. 04 and 46. 65) (type 2 respectively with statistics-0. 1, 1. 24, 0. 23, 1. 54 and 58. 62) has the most accuracy of steady infilterability rate, estimation. Also was observed using topography data as input together with soil physical characteristics can lead to improvement of the estimation accuracy of steady infilterability rate.