هادی فتاحی

The economic assessment of a drilling project heavily relies on forecasting the drilling machine performance. The net cutting rate of a roadheader is a crucial parameter for performance evaluation, and it can be estimated using various methods. In this study, various models were developed to predict roadheader performance and their accuracy and desirability were compared. The models utilized artificial intelligence techniques such as support vector machine (SVM), artificial neural network (ANN), K-fold nearest neighbor (KNN), random forest (RF), and linear regression (LR). The input variables for these models were rock quality designation (RQD) and the return values of the Schmidt hammer R1, R2, and R3. The modeling task was conducted using the Orange data mining software.

The maximum energy consumption of stone cutting machines is one of the important cost factors during the process of cutting construction stones. Accurately predicting and estimating the maximum energy consumption performance of the cutting machine, along with estimating the cutting costs, can help approach the optimal cutting operating conditions to reduce energy consumption and minimize machine depreciation. However, due to the uncertainty and complexity of building stone textures and properties, determining the maximum energy consumption of the device is a difficult and challenging task. Therefore, this paper employs the rock engineering system method to solve the aforementioned problem. To this end, 120 test samples were collected from a marble factory in the Mahalat region of Iran, representing 12 types of carbonate rocks. The input parameters considered for the analysis were the Mohs hardness, uniaxial compressive strength, Young's modulus, production rate, and Schimazek’s F-abrasiveness factors. In the study, 80% of the collected data, equivalent to 96 data points, were utilized to construct the model using the rock engineering system-based method. The obtained results were then compared with other regression methods including linear, power, exponential, polynomial, and multiple logarithmic regression methods. Finally, the remaining 20 percent of the data, comprising 24 data points, were used to evaluate the accuracy of the models. Based on the statistical indicators, namely root mean square error, mean square error, and coefficient of determination, it was found that the rock engineering system-based method outperformed other regression methods in terms of accuracy and efficiency when estimating the maximum energy consumption.

The deformation modulus indicates the degree of deformability of the rock mass in response to any loading and unloading, and it is important because it plays a role in the design of most underground structures. Estimation of this parameter on site is usually done with the help of two tests of plate loading and dilatometer, which is associated with spending a lot of money and time. In addition, due to the presence of discontinuities and cracks in the rock mass, laboratory tests on core samples also face errors. Today, to define a relationship between a parameter and its dependent parameters and to build a model to estimate or predict the chosen parameter, a variety of computational intelligence methods are used, and of course, they also provide favorable results. The purpose of this research is to use these types of algorithms in order to create an efficient model for predicting the deformation modulus on a database. In this regard, the performance of three models created by artificial neural network, K-nearest neighbor and random forest methods have been evaluated with the Orange software. The results showed that the artificial neural network model has the best performance and accuracy with RMSE=0.116 and MAE=0.094. Also, the sensitivity analysis of the input parameters shows that the RMR system is considered as an important and effective parameter.

Compacted soils, which are commonly used in geotechnical engineering projects, such as earth dams, highways, embankments, and airport runways, are mostly unsaturated. To achieve a safe design in all these projects, the stress state variable in soil plays a significant role. Any proposed model for the stress state variable should express its independence from the soil properties. In saturated soils, the e effective stress is taken into account as the stress state variable. Some researchers have attempted to find the stress state variable for unsaturated soils the same as that for saturated soils with only one variable; however, they have noticed that the soil properties have been involved in the proposed models (Bishop, 1959; Escario and Saez, 1986; Khalili and Khabbaz, 1998; Lu and Likos, 2004; Rahnema et al., 2019). The purpose of this paper is to apply new intelligent methods to accurately estimate the effective stress parameter, using two gray wolf optimization (GWO) and sine-cosine (SCA) optimization algorithms.

Determining the lateral bearing capacity of piles is one of the important issues that due to factors such as heterogeneity of soil environment in the pile and its geometric characteristics, the correct estimation of lateral bearing capacity of a pile is a problem. Although pile loading testing can be used as a reliable method at various stages of design despite its high accuracy, it incurs high costs and long execution times for construction and mining projects. In this paper, in order to apply new intelligent methods to estimate the lateral bearing capacity of piles in soft clay soils, two bee colony optimization and invasive weed optimization algorithms have been used.Finally, in order to evaluate the accuracy of the models, the indices of square correlation coefficient (R2), variance inclusion (VAF), mean absolute error percentage (MAPE), root mean square error (RMSE) and mean square error (MSE) were used. The evaluation results showed that for the bee colony algorithm, the estimation accuracy using R2 and VAF indices was about 0.98-0.99 and using MAPE, RMSE and MSE indices between 0.000032-0.0056 and also for the algorithm. For invasive weed optimization algorithm, estimation accuracy using R2 and VAF indices was about 0.97-0.98 and using MAPE, RMSE and MSE indices was between 0.0005-0.023. In addition, the results of sensitivity analysis showed that the non-drained shear strength of soil (Su), among other input parameters, has the greatest impact on the lateral bearing capacity of piles and change in it will have the greatest impact on the output.

Estimation  of  the  load  carrying  capacity  of  pile  foundation  is  one  of  the  most  sought  after  research  areas  in geotechnical   engineering.   Static   equilibrium   and   other dynamic equations are used to predict  the  axial  load  capacity  of  pile.  The  prediction  of lateral  load  capacity  of  piles,  used  in  tall  and  offshore structures is more complex and requires solution of non-linear differential equations. The elastic analysis adopting Winkler   soil   model is   not suitable for the non-linear soil behavior.  Estimating the load capacity of such piles using experimental methods is always associated with error and makes the modeling result far from reality. Today, intelligent methods have shown a high capability in predicting and estimating unknown variables and can replace experimental and analytical methods. In this research, we tried to accurately predict the lateral load capacity of piles in clay soils by creating an intelligent hybrid model called optimized relevant vector regression with the artificial bee colony algorithm. The relevant vector regression is a probabilistic method based on Bayesian approach. The relevant vector regression does not need to predict the error/margin tradeoff parameter C, which can decrease the time and the kernel function, does not need to satisfy the Mercer condition. For those relevant vector regression advantages compared with the support vector regression approach, relevant vector regression model is successfully applied in regression prediction problems. In this method, relevant vector regression is used as a predictive model and artificial bee colony algorithm is used to optimize the parameters of relevant vector regression method. The artificial bee colony algorithm is a swarm based meta-heuristic algorithm for optimizing numerical problems. It was inspired by the intelligent foraging behavior of honey bees. The algorithm is specifically based on the model for the foraging behavior of honey bee colonies. The model consists of three essential components: employed and unemployed foraging bees, and food sources. The first two components, employed and unemployed foraging bees, search for rich food sources, which is the third component, close to their hive. The model also defines two leading modes of behavior which are necessary for self-organizing and collective intelligence: recruitment of foragers to rich food sources resulting in positive feedback and abandonment of poor sources by foragers causing negative feedback.  In artificial bee colony, a colony of artificial forager bees (agents) search for rich artificial food sources (good solutions for a given problem). To apply artificial bee colony, the considered optimization problem is first converted to the problem of finding the best parameter vector which minimizes an objective function. Then, the artificial bees randomly discover a population of initial solution vectors and then iteratively improve them by employing the strategies: moving towards better solutions by means of a neighbor search mechanism while abandoning poor solutions.In this modeling, the data used are related to a laboratory data set of small-scale pile load capacity. Various statistical indicators were used to evaluate the modeling accuracy. Finally, the results showed that the combined relevant vector regression with the artificial bee colony algorithm for test data with R2 = 0.975 and RMSE = 0.001, has a high ability to predict the lateral load capacity of spark plugs. In addition, the sensitivity analysis performed in this study showed that the variables of eccentricity of load and the length of pile are more important and effective compared to other parameters.

One of the indicators for evaluating the performance of a tunnel drilling machine is predicting the penetration rate of this machine. There are various methods and relationships for predicting the penetration rate, each of which has its own characteristics and are presented based on the parameters related to the rock mass and the characteristics of the machine. In this study, genetic, artificial immune system, dolphin echolocation and grey wolf algorithms were used to predict the penetration rate of TBM. In this regard, the database consists of 153 data (122 data for train and 31 data for test) including parameters of intact rock such as strength and brittleness and rock mass characteristics such as distance between planes of weakness and orientation of discontinuities along with TBM machine performance in Queens tunnel has been collected. Mean square error (MSE) and square correlation coefficient (R2) have been used to estimate the error rate between the developed methods. Considering the key parameters of rock mass and intact rock and TBM, relationships to predict the penetration rate are presented and based on statistical analysis, the best relationship is selected. The results are compared with the real data and the results of other models show that the values penetration rate predicted by the genetic algorithm with MSETrain=0.012, MSETest=0.02, R2Train=0.9319 and R2Test=0.8473,has acceptable accuracy compared to other methods.

Due to the difficulties in assessing the deformation of jointed aggregates at the laboratory scale, various in situ testing methods such as plate loading test and dilatometry can be used to consider the effect of scale and joints. Although these methods are currently the best, they are expensive, time consuming, and have operational difficulties during implementation. Therefore, in this paper, to overcome these problems, new harmony search algorithms (HS) and teaching-learning optimization algorithm (TLBO) are used to indirectly estimate the modulus of rock mass deformation. In these models, the rock mass classification score (RMR), uniaxial compressive strength of virgin rock (UCS), depth (D) and the modulus of elasticity of intact rock (Ei) as input parameters and the modulus of rock mass deformability (Em) as output parameter Used. In this paper, Using different statistical indicators, the model created by the algorithms is evaluated and validated. The evaluation results showed that the relationship accuracy for the harmonic search algorithm using R2 and VAF methods is about 0.91-0.93 and using the RMSE and MSE methods is between 0.000017-0.0042. Also, the relationship accuracy for the optimization algorithm Based on teaching and learning using R2 and VAF methods, about 0.92-0.95 and using RMSE and MSE methods were between 0.00001- 0.0032.

The purpose of grouting is to strengthen and improve the mechanical and hydraulic properties of the rock and soil. The fluid that is injected into the cavities and fissures of the environment is like a viscous liquid consisting of grains whose size is important in the grouting operation. Therefore, determining the groutability ratio in grouting operation is considered as an important parameter. Today, studies using data mining science show that the groutability of granular soils, in addition to grain size, is affected by various factors of the soil and the material of grout, which predicts groutability more accurately. Throughout history, many researchers have predicted groutability through experimental relationships. However, today, the capability of data mining methods in accurate predictions has shown that one approach in predicting groutability is to use a variety of data mining models and inferential systems.

Methodology and Approaches:

The purpose of this study is to evaluate several models of data mining methods, including ANN, SVM, KNN, RF and NB. For this purpose, a set of laboratory information related to groutability has been used in four literatures that include 87 data in order to develop efficient models for predicting groutability. Classification models are created in Orange software.

The output variable is a property of groutability, which as a binary variable has two states of zero meaning nongroutable and 1 meaning groutable. Input variables also include the ratio of water cement in the grout or viscosity (W/C), the relative density of the soil (Dr), grouting pressure (P), the percentage of the soil particles passing through a 0.6 mm sieve (FC), N1 = D15soil / D85 grout and N2 = D10 soil / D95 grout. The values of the evaluation criteria for the methods are almost close to each other. Based on the AUC index, the random forest is the best model and the k-nearest neighbor method has the lowest value of this index. However, in terms of other criteria, the artificial neural network is higher than other methods and the k-nearest neighbor method is very close to it. On the other hand, the random forest model has the lowest value of criteria. Ignoring the AUC criteria, ANN and KNN methods are the best methods.One of the capabilities of Orange software is to study the effect and importance of input variables on the prediction of the target variable, in other words, the sensitivity of the output variable to input variables. The results show that variable N2 is in the first level based on the three criteria of information gain, relative information gain and Gini index, and variable N1 is in the second level with a very small difference in the values of the criteria. In addition, in the last row, W/C has the lowest value of the criteria and shows a small role in the correct prediction of groutability.

In predicting the behavior of geotechnical structures that are constructed on unsaturated soils, knowing the geomechanical properties, especially effective stress, is of particular importance. The effect of suction in unsaturated soils changes the volumetric and shear behavior of soil. Therefore, in determining the shear strength in unsaturated soils, the effective stress parameter (χ) plays an essential role as a function of suction. Determining this parameter requires a lot of time and money that is spent on laboratory tests. The purpose of this study is to evaluate several intelligent methods for creating models that indirectly accurately estimate parameter (χ). For this purpose, 120 data (which are the results of thriaxial, shear, pressure plate and filter paper tests) and intelligent methods of random forest, support vector machine and k-nearest neighbor were used in WEKA intelligent software. The results show that the models developed by the three intelligent methods of random forest, support vector machine and k-nearest neighbor, have excellent performance and accuracy. But the random forest model is better than the other two models in evaluating the test data with R2 = 0.918 and RMSE = 0.079. In addition, in this study, sensitivity analysis was performed to determine the importance of the parameters affecting parameter (χ). Among the input parameters in modeling, it was found that the volume water content parameter (θ_r/θ_s ) has the greatest effect on parameter (χ).

Probabilistic slope stability analysis provides a tool for considering uncertainty of the soil parameters in design. In this paper, using Slide software, the Monte Carlo simulation is used as an analytical method to develop probabilistic models of slope stability based on equilibrium methods. The limit equilibrium methods are the most popular approaches in slope stability analysis. These methods are well known to be a statically indeterminate problem, and assumptions on the inter-slice shear forces are required to render the problem statically determinate. In modeling using this methodology, the selected stochastic parameters are internal friction angle, cohesion and unit weight of soil, which are modeled using a truncated normal probability density function (pdf). In this research, the abilities offered using models were presented by using field data obtained from Cannon dam in Iran. The results obtained show that the hybrid Monte Carlo simulation and equilibrium methods can be used successfully for slopes stability assessment.

With the increasing expansion of urban environments, the creation and development of intra-city transportation systems in order to reduce traffic, pollution and reduce the costs of intra-city traffic is essential. Considering that an important part of the construction cost of the metro is related to the excavation and maintenance of tunnels. Therefore, one of the most important decisions in the construction of subway tunnels is the excavation method in alluvial and fall environments. Tunnel excavation by TBM-EPB machine is a fast, powerful and maintenance method compared to other excavation methods in soft soils and fall areas. One of the most important factors in preventing the face pressure from falling during excavation in soft and alluvial fields is estimating the optimal face pressure of the excavation machine in each excavation stage (different kilometers). Because the high or low face pressure of the excavation machine leads to increased costs, loss of life, high hardness and also leads to delays in the completion of the project. In this paper, due to the uncertainty in geotechnical parameters and the sensitivity of urban tunnels, the issue has been studied from a probabilistic perspective. For this purpose, first for 50 different numerical modeling modes of Tabriz Metro Line 2 using PLAXIS3D2020 software and then Monte Carlo simulation method has been used to generate random numbers and assign appropriate probabilistic distributions. Then, using the Gray Wolf meta-heuristic algorithm (GWO), the face pressure of the TBM-EPB machine was estimated using the prediction relation. Finally, in order to evaluate and validate the relationship, the statistical indicators of square correlation coefficient (R2), variance inclusion (VAF), mean absolute error percentage (MAPE), root mean square error (RMSE) and mean square error (MSE) were used. Is. According to the model validation, the relationship created by the gray wolf algorithm is very close to the reality of the problem and it can be used to continue the route in other similar areas.

With the development and growth of the population in big cities; The need to use public transportation facilities such as the metro is growing. Given the importance of the urban transportation network, it is very important to pay attention to the health control of underground structures against static and dynamic loads. As Iran is one of the seismic countries in the world, in recent years it has witnessed devastating earthquakes and has caused heavy damage. According to previous studies, underground structures are more safe against earthquakes. Because surface structures are connected to the ground only on the floor and lower surface. If the underground structures are in full conflict with the surrounding environment and are more resistant to earthquake loads. However, given that most subways are built on shallow urban soils, there have been reports of severe damage to these underground spaces from earthquake loads. For this reason, identifying damage and controlling the health of underground spaces, especially subways, is very important. In this research, due to the uncertainty in the ground parameters and also the lack of accurate forecast of the drilling route, 3DEC software (when dealing with rock layers) along with PLAXIS3D2020 software has been used for static and dynamic analysis of Mashhad Metro Line 2 tunnel. In addition, in order to control the health of the structure (Mashhad Metro Line 2), wavelet transform (WT) methods have been used. For this purpose, the received signal was called by dynamic analysis and using the wavelet conversion toolbox in MATLAB software, damage areas were identified in the model (around the concrete cover and side borders), which obviously can be identified using the maintenance system. Properly prevented the collapse of the underground structure in the identified areas.

Roadheader machine is one of those machines that have high drilling capability in rocks with low to medium strength. Hence they are widely used in underground excavations. Estimating the performance of roadheader machine is one of the main and important issues in estimating the approximate project completion time as well as project costs. Therefore, the purpose of this paper is to propose intelligent forecasting models for estimating the performance of roadheader machine by two intelligent methods (the firefly algorithm (FA) and the Teaching-learning based optimization algorithm (TLBO)) and using a database (a case study). Is. In these models, the Schmidt hammer rebound values and the rock quality degree (RQD) are used as input parameters and the cutting rate of the roadheader is used as the output parameter. Finally, to evaluate the accuracy of the models and modeling, the indices of square correlation coefficient (R2), variance account for (VAF), root mean square error (RMSE) and mean square error (MSE) have been used. The results indicated that the two models have strong potentials to estimate roadheader performance with high degrees of accuracy and robustness.

One of the important issues in designing non-level tunnels is determining the optimal location of tunnels relative to each other. In this research, to optimally locate the two factors of land subsidence and stability (safety factor) of tunnels have been used. Optimal placement of non-level tunnels in various geological conditions should use probabilistic methods and reliability. In this paper, PLAXIS3D finite element software is used to study different states of non-level tunnels. Then, using the harmony search algorithm, two limit state functions are estimated separately for the safety factor and maximum ground surface subsidence. Then, using the two limit state functions obtained from the previous step and the first-order reliability methods and the Monte Carlo simulation in RT software, the best location for excavation of the second tunnel in relation to the previous (existing) tunnel is based on having the highest safety factor and lowest ground subsidence. 

Due to the complexity of the interaction between the tunnels and also due to the uncertainty in the design parameters, in this paper, using probabilistic analyzes such as the Monte Carlo simulation method and first-order reliability, the optimal stability and placement of non-level tunnels in RT probabilistic software was discussed. Because RT software requires a correct and accurate limit state function, the harmonic search algorithm in MATLAB was used to calculate the stability and also to achieve this limit state function. 

In this paper, using PLAXIS3D software, the tunnel was modeled in 32 different rock masses and the results are different in each of the 32 rock masses due to the inequality in the input parameters (safety factor and maximum settlement). To evaluate the reliability of the new tunnel excavation (existing sub-tunnel), 32 models which were analyzed by PLAXIS3D software were compared with the predicted model to achieve the lowest settlement rate and the highest safety factor by the harmony search algorithm. The comparison showed good consistency between the model predicted by the harmony search algorithm and the model performed by numerical methods. Therefore, due to the proximity of the performed model and the predicted model, the limit state function is accurate to obtain the probability of failure of the new tunnel. 

According to the results obtained in both Monte Carlo simulation methods and first-order reliability, the probability of failure (approximately 0.33%) and the high-reliability index are shown and the tunnel position is in very good condition. According to the analysis of tunnel random variables by RT software, it was found that the geological durability index is very important and effective. It should be noted that the results of this study show that the reliability methods used in the RT program to stabilize and locate the new tunnel under the existing tunnel can be used as a high-performance method in analyzing underground space problems.

One of the possible consequences of earthquakes in saturated areas is soil liquefaction and as a result the failure of foundations of buildings, types of infrastructure, bridges and many other disasters. In this study, in order to evaluate the potential of soil liquefaction on 79 samples from China Tangshan Earthquake Database, several intelligent classification models were constructed with the help of Orange software. Therefore, the performance of 5 intelligent classification methods (Logistic Regression, Artificial Neural Network (ANN), Support Vector Machine (SVM), K-fold Nearest Neighbor (KNN) and Random Forest) were compared based on different criteria. The results showed that SVM, ANN and Logistic Regression methods have a high ability to predict soil liquefaction class and among them the Logistic Regression method with AUC index (0.98) was selected as the best method. In addition, the study of the effectiveness of variables using three criteria of Information Gain, Information Gain Ratio and Gini Index, indicates that the variable measured CPT tip resistance is the most effective variable and is the first priority. The variables of cyclic stress ratio and peak acceleration at the ground surfaceare also important features.

The surface settlement is an essential parameter in the operation of mechanized tunneling that should be determined before excavation. The surface settlement analysis caused by mechanized tunneling is a geo-technical problem characterized by various sources of uncertainty. Unlike the deterministic methods, the reliability analysis can take into account the uncertainties for the surface settlement assessment. In this work, the reliability analysis methods (second-order reliability method (SORM), Monte Carlo simulation (MCS), and first-order reliability method (FORM)) based on the genetic algorithm (GA) are utilized to build models for the reliability analysis of the surface settlement. Specifically, for large-scale projects, the limit state function (LSF) is non-linear and hard to apply based on the reliability methods. In order to resolve this problem, the GMDH (group method of data handling) neural network can estimate LSF without the need for additional assumptions about the function form. In this work, the GMDH neural network is adapted to obtain LSF. In the GMDH neural network, the tail void grouting pressure, groundwater level from tunnel invert, depth, average penetrate rate, distance from shaft, pitching angle, average face pressure, and percent tail void grout filling are used as the input parameters. At the same time, the surface settlement is the output parameter. The field data from the Bangkok subway is used in order to illustrate the capabilities of the proposed reliability methods.

In recent years, along with the development and expansion of cities, population growth and vehicles, the construction of underground railways by mechanized tunneling is increasing. One of the common problems of EPB-TBM mechanized tunneling in weak and unstable areas is tunnel convergence, which causes problems such as getting stuck on EPB-TBM, subsidence and damage to buildings around the tunnel in urban areas. Due to the uncertainty in the soil parameters of the region, estimating the face pressure, thrust jacks pressure and cutter head torque at different kilometers is very important to prevent over-convergence of the tunnel, getting stuck on EPB-TBM and ground subsidence. Due to the uncertainty in rock and soil, in this paper the amount of face pressure, thrust jacks force, cutter head torque and ground subsidence due to tunneling in different kilometers in the construction project of the new Istanbul airport in Turkey using numerical software, PLAXIS3D 2020, is modeled. The modeling results show that the operator can increase the face pressure by preventing getting stuck on EPB-TBM. According to the results obtained in this paper, adjusting the face pressure, thrust jacks force and cutter head torque, in addition to solving the problem of getting stuck on EPB-TBM, will reduce ground subsidence and damage structures on the ground surface.

The uniaxial compressive strength parameter is an important and key parameter in geomechanical engineering that is used in rock engineering classification, study of rock fracture criteria and in the design stage of many construction and mining projects. In many cases, due to the lack of access to quality cores, determining these parameters in the laboratory is associated with many difficulties, and often this parameter is indirectly estimated from regression relationships that these relationships are not very accurate. The aim of this study is to use the new meta-heuristic algorithms of gray wolf (GWO) and fruit fly algorithm (FFOA) to indirectly estimate uniaxial compressive strength. To achieve this goal, data from 124 granite rock samples from the Selangor freshwater transfer tunnel project in Malaysia, have been used. Finally, to evaluate and validate the models obtained by intelligent algorithms, the indices of statistical are used. According to the results obtained in this paper as well as the validation of the models, the predicted values ​​of uniaxial compressive strength are very close to the real values ​​of the region, indicating the low error of the models in indirect estimation. In addition, in this paper, sensitivity analysis was performed on the effective parameters in estimating uniaxial compressive strength. The results showed that the return values ​​of Schmidt hammer (Rn), among other input parameters, have the greatest effect on uniaxial compressive strength.

Rockburst in deep underground excavations is a phenomenon that is observed in the form of sudden rock failure and release of strain energy stored in underground mines and rock tunnels, usually in stressful places and at a depth greater than the earth's surface. Due to this explosive failure, rock in pieces Small and large are scattered around and cause damage to humans or equipment. In the present study, the potential of rockburst using data mining techniques and comparing their results with three experimental methods, tangential stress criterion (SC), brittleness criterion (BC) and elastic energy index (EEI) are investigated. For this purpose, data mining models such as support vector machine (SVM), k-nearest neighbor (KNN), Bayesian networks (BNs), artificial neural network (ANN) and CHAID tree model have been used in WEKA software. The results show the superiority of data mining algorithms over experimental methods. Among data mining algorithms, the support vector machine model had the highest accuracy with 80.8% accuracy. However, measuring the accuracy of the models by RMSE method considers the artificial neural network model with the value of RMSE = 0.337 as the best model. In addition, among the experimental methods, the elastic energy index method with an accuracy of 60.7 is the best method.

Rock fall is a type of slope instability that occurs in most mountainous areas and causes casualties and severe damage to facilities, residential areas, roads, farms and vehicles. Therefore, to prevent financial damage, it is necessary to predict and model the phenomenon of rock fall in steep walls to take preventive measures. In this paper, in order to assess the risk of falling rocks, first using Rocfall software, the path of rock fall of isolated blocks was analyzed individually and separately in the most critical section of the northwestern heights of Shahrood. In order to consider the interactions between particles in the path, shape, size and weight of the parts, the distinct element method PFC3D was used. Analyzing the simulation results with the help of these two softwares, it was concluded that if the rock falls, a large number of parts will stop in the area of ​​roads and residential areas, which will result in traffic jams and financial-human risks. In other words, considering that the maximum horizontal distance traveled by the plots is 393 meters and the horizontal distance between residential areas is 340 meters, so the probability of an accident due to falling rocks is very high. In the following, suggestions such as installation of metal grids, shock absorbers, reduction of kinetic energy in the parts, change in the dimensions of the parts, change of the slope geometry to reduce the risk and damage caused by falling rock in the desired section are presented.

The mechanical characteristics of rocks and rock masses are considered as the determining factors in making plans in the mining and civil engineering projects. Two factors that determine how rocks responds in varying stress conditions are P-wave velocity (PWV) and its isotropic properties. Therefore, achieving a high-accurate method to estimate PWV is a very important task. This work investigates the use of different intelligent models such as multivariate adaptive regression splines (MARS), classification and regression tree (CART), group method of data handling (GMDH), and gene expression programming (GEP) for the prediction of PWV. The proposed models are then evaluated using several error statistics, i.e. squared correlation coefficient (R2) and root mean squared error (RMSE). The values of R2 obtained from the CART, MARS, GMDH, and GEP models are 0.983, 0.999, 0.995, and 0.998, respectively. Furthermore, the CART, MARS, GMDH, and GEP models predict PWV with the RMSE values of 0.037, 0.007, 0.023, and 0.020, respectively. According to the aforementioned amounts, the models presented in this work predict PWV with a good performance. Nevertheless, the results obtained reveal that the MARS model yields a better prediction in comparison to the GEP, GMDH, and CART models. Accordingly, MARS can be offered as an accurate model for predicting the aims in other rock mechanics and geotechnical fields.

Dams are mainly constructed of earth and rock-fill materials and hence they are generally referred to as embankment dams orfill-type dams. The role of drainage system is also vital as it shifts the phreatic surface ensuring the safety of downstreamtoe. This paper presents the results of seepage analyses of the considered earth dam using finite element method. Slide software is one of geotechnical program that is based on the finite element and canconsider analysis like stress-strain, seepage, slope stability, dynamic analysis and also fast water drop inreservoir. In this research seepage analysis in Roudbar Lorestan dam has been done by Slide software. In orderto evaluate the type and size of mesh size on the total flow rate and total head through the dam cross section,four mesh size such as coarse, medium, fine and unstructured mesh is considered. Result showed that averageflow rate of leakage under the different mesh size for Roudbar Lorestan dam equal 629584 m3 per year for the entire lengthof the dam.

Probabilistic slope stability analysis provides a tool for considering uncertainty of the soil parameters in design. In this paper, using Slide software, the Monte Carlo simulation is used as an analytical method to develop probabilistic models of slope stability based on equilibrium methods. The limit equilibrium methods are the most popular approaches in slope stability analysis. These methods are well known to be a statically indeterminate problem, and assumptions on the inter-slice shear forces are required to render the problem statically determinate. In modeling using this methodology, the selected stochastic parameters are internal friction angle, cohesion and unit weight of soil, which are modeled using a truncated normal probability density function (pdf). In this research, the abilities offered using models were presented by using field data obtained from Roudbar Lorestan dam in Iran. The results obtained show that the hybrid Monte Carlo simulation and equilibrium methods can be used successfully for slopes stability assessment.

Study of immediate settlement of a rectangular foundation with the task of transferring structural forces to the ground, is one of the important factors in the design of structures. Due to its heterogeneous and anisotropicnature, soil parameters have highuncertainties, the omission of which in estimating the settlement of foundation, creates challenges in the analysis of structural stability. Therefore, the use of reliability methods with the ability to accept uncertainties and reliable stability analysis of geotechnical issues are welcomed today. In this study, in order to estimate the probability of failure due to settlement of rectangular foundation, first-order and second-order reliability methods as well as Monte Carlo simulations were used and the relevant analysis were performed in RT and @Risk software.The results indicate a high probability of failure related to the foundation settlement (9.5%). The second-order methods and Monte Carlo simulations provide more accurate results due to the non-linearity of the limit state procedure. In addition, the sensitivity analysis of random variables based on sensitivity vectors showed that thecontact stress (qo), has an important role in the probability of failure, and then the Poisson’s ratio is important.