Search

Tight carbonate reservoirs appear to be heterogeneous due to the patchy production of various digenetic properties. Consequently, the permeability calculation of tight rocks is costly, and only a finite number of core plugs in any single reservoir can be estimated. Hence, in the present study, a novel hybrid model constructed by combination of the improved version of the Harris Hawks optimisation (HHO), i.e., IHHO, and extreme learning machine (ELM) is proposed to predict the permeability of tight carbonates using limited number of input variables. The proposed IHHO employs a mutation mechanism to avoid trapping in local optima by increasing the search capabilities. Subsequently, ELM-IHHO, a novel metaheuristic ELM-based algorithm, was developed to predict the permeability of tight carbonates. Experimental results show that the proposed ELM-IHHO attained the most accurate prediction with R2 = 0.9254 and RMSE = 0.0619 in the testing phase. The result of the proposed model is significantly better than those obtained from other ELM-based hybrid models developed with particle swarm optimisation, genetic algorithm, and slime mould algorithm. The results also illustrate that the proposed ELM-IHHO model outperforms the other benchmark model, such as back-propagation neural nets, support vector regression, random forest, and group method of data handling in predicting the permeability of tight carbonates.

A proper and reliable estimation of bearing capacity of thin-walled foundations is of importance and necessary for accurate design of these structures. This study proposes a new hybrid intelligent technique, i.e., adaptive neuro-fuzzy inference system (ANFIS)–polynomial neural network (PNN) optimized by the genetic algorithm (GA), called ANFIS–PNN–GA, for prediction of bearing capacity of the thin-walled foundations. In fact, in ANFIS–PNN–GA system, GA was used to optimize the ANFIS–PNN structure. To achieve the aim of this study, a series of data samples were collected from literature. After establishing the database, many ANFIS–PNN–GA models were constructed and proposed to estimate the bearing capacity of the aforementioned foundations. To show capability of this advance hybrid model, two pre-developed models i.e., ANFIS and PNN were also built to predict bearing capacity. The performance prediction of the proposed models were evaluated through the use of several performance indices, e.g., correlation coefficient (R) and mean square error (MSE). The R values of (0.9825, 0.9071, and 0.9928) and (0.8630, 0.7595 and 0.9241) were obtained for training and testing data of the ANFIS, PNN and ANFIS–PNN–GA, models, respectively. Accordingly, because of the role of GA as a practical optimization algorithm in improving the efficiency of both PNN and ANFIS models, results obtained by the ANFIS–PNN–GA model are more accurate compared to other implemented methods. The proposed advance hybrid model can be introduced as a new and applicable technique for solving problems in field of geotechnics and civil engineering.

Thermal conductivity is a specific thermal property of soil which controls the exchange of thermal energy. If predicted accurately, the thermal conductivity of soil has a significant effect on geothermal applications. Since the thermal conductivity is influenced by multiple variables including soil type and mineralogy, dry density, and water content, its precise prediction becomes a challenging problem. In this study, novel computational approaches including hybridisation of two metaheuristic optimisation algorithms, i.e. firefly algorithm (FF) and improved firefly algorithm (IFF), with conventional machine learning techniques including extreme learning machine (ELM), adaptive neuro-fuzzy interface system (ANFIS) and artificial neural network (ANN), are proposed to predict the thermal conductivity of unsaturated soils. FF and IFF are used to optimise the internal parameters of the ELM, ANFIS and ANN. These six hybrid models are applied to the dataset of 257 soil cases considering six influential variables for predicting the thermal conductivity of unsaturated soils. Several performance parameters are used to verify the predictive performance and generalisation capability of the developed hybrid models. The obtained results from the computational process confirmed that ELM-IFF attained the best predictive performance with a coefficient of determination = 0.9615, variance account for = 96.06%, root mean square error = 0.0428, and mean absolute error = 0.0316 on the testing dataset (validation phase). The results of the models are also visualised and analysed through different approaches using Taylor diagrams, regression error characteristic curves and area under curve scores, rank analysis and a novel method called accuracy matrix. It was found that all the proposed hybrid models have a great ability to be considered as alternatives for empirical relevant models. The developed ELM-IFF model can be employed in the initial stages of any engineering projects for fast determination of thermal conductivity.

A proper planning schedule for tunnel boring machine (TBM) construction is considered as a necessary and difficult task in tunneling projects. Therefore, prediction of TBM performance with high degree of accuracy is needed to prepare a suitable planning schedule. This study aims to predict the advance rate of TBMs using optimized extreme learning machine (ELM) model with six particles swam optimization (PSO) techniques. Hence, six deterministically adaptive models, including time-varying acceleration (TAC)–PSO–ELM, improved PSO–ELM, Modified PSO–ELM, TAC–MeanPSO–ELM, improved MeanPSO–ELM, and Modified MeanPSO–ELM were developed. A number of performance criteria along with ranking system were used to identify the best model. The results showed that modified MeanPSO–ELM achieved the highest cumulative ranking (56), while the modified PSO–ELM achieved the lowest cumulative ranking (51). For training phase, improved PSO–ELM and TAC–PSO–ELM achieved the highest ranking (30) for each. The TAC–MeanPSO–ELM obtained the lowest ranking in the testing phase (29). Concerning the coefficient of determination (R2), modified PSO–ELM, improved PSO–ELM, TAC–PSO–ELM, and modified MeanPSO–ELM showed a similar behavior and achieved 0.97 for training and 0.96 for testing phases. Two models, including improved MeanPSO–ELM and TAC–MeanPSO–ELM achieved the same R2 of 0.96 for both training and testing phases. The findings of this study suggest that the hybridization of ELM and PSO may result in more accurate results than single ELM model to predict the TBM advance rate.

The occurrence of unpredictable hazards are frequent with the increased depth of mining, especially the hazards caused by stress concentration. In order to mitigate the negative effectiveness results from mining-induce stress, various approaches have been employed in underground mines. Destress blasting, as an efficient method, has gained a lot of popularity in recent years. However, it is crucial to estimate the destressability of specific area before conducting destress blasting. In this study a combination model on the basis of both unascertained measurement (UM) and entropy coefficients was applied to observe the performance of destressability evaluation. Eight representative parameters, i.e., stiffness of the rock mass, brittleness of the rock mass, degree of fracturing, proximity to failure, destress blast orientation, width of the target zone, unit explosive energy, and confinement of the charges were chosen as initial input parameters, and their membership distributions were described by four types of membership methodologies, i.e., line, parabolic curve, exponential curve, and sine curve. Meanwhile, the weights of each index could be computed based on the single measurement matrix. Then, destressability of the samples was easily identified with Euclidean distance and comprehensive measurement vectors which were computed by single measurement vectors and weight coefficients. Finally, it was found that the assessment results are in accordance with those calculated by destressability index. It can be concluded that the proposed hybrid model is able to eliminate the disturbance of subjective factors and ensure the reliability of these outcomes. At the same time, it can provide a novel idea/process for the destressability evaluation.

The relevance vector machine (RVM) is considered a robust machine learning method and its superior performance has been confirmed through many successful engineering applications. To improve the performance of the RVM model, three single kernel functions, and three multikernel functions, including two newly proposed multikernel functions, tenfold cross-validation, and the hybrid particle swarm optimization with grey wolf optimizer (HPSOGWO) algorithm were combined to develop an artificial intelligence (AI) model framework. Afterwards, a new application of the RVM method was used and introduced for two different datasets of the blast-induced ground vibration. In addition, an artificial neural network (ANN) model and seven empirical equations were also developed for comparison purposes, and their prediction performances were evaluated considering three performance metrics, i.e., root mean square error (RMSE), correlation coefficient (R2), and mean absolute error (MAE). The obtained results showed that the multikernel RVM model can provide better performance capacity than the single-kernel RVM model. As a result, the AI models were found to be more applicable than the empirical equations in estimating blast-induced ground vibration. The prediction performance results of these models confirmed that the selected database has a great impact on the prediction capacity. Therefore, it is a common act to compare the performance of various models based on the selected database before selecting an optimal predictive model. The proposed model in this study provides new theoretical and practical support for the prediction of blast-induced ground vibration and can be utilized by other researchers in similar fields.

The aim of this paper is to extend a stochastic model for simulation of highly nonstationary ground motions (GMs). In this model, the dual-tree complex discrete wavelet transform (DT-CDWT) is applied to a recorded ground motion to extract its wavelet coefficients and then the Smooth Transition Generalized Autoregressive Conditional Heteroscedastic (ST-GARCH) model is used to simulate these coefficients. This model simulates nonstationary features of real GMs greatly, because the conditional variance estimated by this model changes compatibly with the amplitude nonstationary features of real GMs. Because of having asymmetric structure, this model also estimates sudden changes in the amplitude of ground motions. Some of the special capabilities of this model are the simulation of multiple increasing–decreasing cycles in the temporal amplitude of GMs, prediction of several peaks in the frequency spectrum of GMs, estimation of steps of their energy curve, and simultaneous simulation of all shocks of a GM sequence. Also, this model simulates the time–frequency energy distribution of both wide-frequency and narrow-frequency bandwidth GMs.

Air overpressure (AOp) induced by rock blasting is an undesirable phenomenon in open-pit mines and civil construction works. The prediction of AOp has been always a complicated task since many parameters have potential to affect the propagation of air waves. This study aims to assess the capability of a new hybrid evolutionary model based on an integrated adaptive neuro-fuzzy inference system (ANFIS) with a stochastic fractal search (SFS) algorithm. To assess the reliability and acceptability of ANFIS-SFS model, the particle swarm optimization (PSO) and genetic algorithm (GA) were also combined with ANFIS. The proposed models were developed using a comprehensive database including 62 sets of data collected from four granite quarry sites in Malaysia. Performances of the ANFIS-SFS, ANFIS-GA, and ANFIS-PSO models were checked using statistical functions as the performance criteria. The obtained results showed that the proposed ANFIS-SFS model, with root mean square error of 1.223 dB, provided much higher generalization capacity than the ANFIS-PSO (RMSE of 1.939 dB), ANFIS-GA (RMSE of 2.418 dB), and ANFIS (RMSE of 3.403 dB) models in terms of predicting AOp. This clearly demonstrates the effectiveness of SFS to provide a more accurate model in the AOp prediction field.

In geotechnical engineering, excavation process is affected by many factors, so that the evaluation of the excavatability is a difficult task in the field. In order to have a better equipment selection and minimize the excavation cost, it seems that there is a need to develop a more accurate evaluation model for excavatability. In this paper, a hybrid model based on unascertained measurement and information entropy is proposed to assess the excavatability of rock mass in specific area. The developed model differs from other available evaluation methods; the former is able to keep objective in the whole calculation process. In order to achieve aims of this study, five groups of data including cuttability, rippability, drillability, blastability and diggability were collected from the literature to test the performance of the developed model. The single-index measurement values were calculated through the linear, parabolic, exponential and sine membership functions, and the weights of evaluation indices were determined by information entropy theory. Besides, the criterion of credible degree (Rec) ranging from 0.5 to 0.7 was able to provide some information influenced by different Rec to the evaluation results. Findings revealed that the hybrid model of unascertained measurement and information entropy can be introduced as an accurate and applicable tool in the field of excavation as it showed a high efficient level in model assessment.

Effective prediction of the peak shear strength (PSS) is of crucial importance in evaluating the stability of a rock slope with interlayered rocks and has both theoretical and practical significance. This paper offers two novel prediction tools for the PSS prediction based on radial basis function neural network (RBFNN) and meta-heuristic computing paradigms. For this work, the gray wolf optimization (GWO) and ant colony optimization (ACO) algorithms were used to select the optimal parameters of RBFNN. Then, these two new models were compared with the gene expression programming (GEP) model. A total of 158 experimental data were used to train and test the proposed models using three input parameters, i.e., normal stress, compressive strength ratio of joint walls, and joint roughness coefficient. Finally, the computational result revealed that the RBFNN-GWO model, with the coefficient of determination (R2) of 0.997, produced a better convergence speed and higher accuracy compared with RBFNN-ACO and GEP models, with the R2 of 0.995 and 0.996, respectively. The RBFNN-GWO model was found an efficient predictive tool that can help rock engineers in the slopes design processes.

To test the impact of different mixture ratios on backfilling strength in Fankou lead–zinc mine, various mixture ratio designs have been conducted. Meanwhile, to improve the strength of ultra-fine tailings-based cement paste backfill (CPB), two kinds of fibers were utilized in this study, namely polypropylene (PP) fibers and straw fibers. To achieve these, a total of 144 CPB backfilling scenarios with different combinations of influenced factors were tested by uniaxial compressive tests. The test results indicated that polypropylene fibers improve the strength of CPB, while in some scenarios the addition of straw fibers decreases the strength of CPB. In this research, the support vector machine (SVM) technique coupled with three heuristic algorithms, namely genetic algorithms, particle swarm optimization and salp swarm algorithm (SSA), was developed to predict the strength of fiber-reinforced CPB. Also, the optimal performance of metaheuristic algorithms was compared with one fundamental search method, i.e., grid search (GS). The overall performance of four optimal algorithms was calculated by the ranking system. It can be found that these four approaches all presented satisfactory predictive capability. But the metaheuristic algorithms can capture better hyper-parameters for SVM prediction models compared with GS-SVM method. The robustness and generalization of SSA-SVM methods were the most prominent with the R2 values of 0.9245 and 0.9475 for training sets and testing sets. Therefore, SSA-SVM will be recommended to model the complexity of interactions for fiber-reinforced CPB and predict fiber-reinforced CPB strength.

Ground vibration is one of the most important undesirable effects induced by blasting operations in the mining or tunneling projects. Hence, developing a precise model for prediction of ground vibration would be much beneficial to control environmental issues of blasting. The present study proposes a new hybrid machine learning (ML) technique, i.e., autonomous groups particles swarm optimization (AGPSO)–extreme learning machine (ELM) to predict ground vibration resulting from blasting. In fact, AGPSO–ELM model is a modified version of PSO–ELM that can solve problems in a way with higher prediction performance. For comparison purposes, PSO–ELM, minimax probability machine regression, least square–support vector machine and Gaussian process regression models were also proposed to estimate ground vibration. The said ML models were trained and tested based on a database comprising of 102 datasets collected from a quarry site in Malaysia. In the modeling of ML techniques, six input parameters were considered: burden to spacing ratio, maximum charge per delay, stemming, distance from the blasting-face, powder factor and hole depth. The results of ML techniques were evaluated in both stages of training and testing based on five fitness parameters criteria. Considering results of both training and testing datasets, AGPSO–ELM model was able to provide higher prediction performance for PPV prediction. Root-mean-square error values of (0.08 and 0.08) and coefficient of determination values of (0.92 and 0.90) were obtained, respectively, for training and testing datasets of AGPSO–ELM model which revealed that the new hybrid model is capable enough to forecast ground vibration induced by blasting. The newly proposed model can be used in other fields of science and engineering in order to get high accuracy level of prediction.

This study proposes a novel design to systematically optimize the parameters for the adaptive neuro-fuzzy inference system (ANFIS) model using stochastic fractal search (SFS) algorithm. To affirm the efficiency of the proposed SFS-ANFIS model, the predicting results were compared with ANFIS and three hybrid methodologies based on ANFIS combined with genetic algorithm (GA), differential evolution (DE), and particle swarm optimization (PSO). Accurate prediction of uniaxial compressive strength (UCS) is of great significance for all geotechnical projects such as tunnels and dams. Hence, this study proposes the use of SFS-ANFIS, GA-ANFIS, DE-ANFIS, PSO-ANFIS, and ANFIS models to predict UCS. In this regard, the fresh water tunnel of Pahang–Selangor located in Malaysia was considered and the requirement data samples were collected. Different statistical metrics such as coefficient of determination (R2) and mean absolute error were used to evaluate the models. Referring to the efficiency results of SFS-ANFIS, it can be found that the SFS-ANFIS (with the R2 of 0.981) has higher ability than PSO-ANFIS, DE-ANFIS, GA-ANFIS, and ANFIS models in predicting the UCS.

This study aims to identify the suitability of hybridizing the firefly algorithm (FA), genetic algorithm (GA), and particle swarm optimization (PSO) with two well-known data-driven models of support vector regression (SVR) and artificial neural network (ANN) to predict blast-induced ground vibration. Here, these combinations are abbreviated using FA–SVR, PSO–SVR, GA–SVR, FA–ANN, PSO–ANN, and GA–ANN models. In addition, a modified FA (MFA) combined with SVR model is also proposed in this study, namely, MFA–SVR. The feasibility of the proposed models is examined using a case study, located in Johor, Malaysia. Then, to provide an objective assessment of performances of the predictive models, their results were compared based on several well known and popular statistical criteria. According to the results, the MFA–SVR with the coefficient of determination (R2) of 0.984 and root mean square error (RMSE) of 0.614 was more accurate model to predict PPV than the PSO–SVR with R2 = 0.977 and RMSE = 0.725, the FA–SVR with R2 = 0.964 and RMSE = 0.923, the GA–SVR with R2 = 0.957 and RMSE = 1.016, the GA–ANN with R2 = 0.936 and RMSE = 1.252, the FA–ANN with R2 = 0.925 and RMSE = 1.368, and the PSO–ANN with R2 = 0.924 and RMSE = 1.366. Consequently, the MFA–SVR model can be sufficiently employed in estimating the ground vibration, and has the capacity to generalize.

Over the last decade, application of soft computing techniques has rapidly grown up in different scientific fields, especially in rock mechanics. One of these cases relates to indirect assessment of uniaxial compressive strength (UCS) of rock samples with different artificial intelligent-based methods. In fact, the main advantage of such systems is to readily remove some difficulties arising in direct assessment of UCS, such as time-consuming and costly UCS test procedure. This study puts an effort to propose four accurate and practical predictive models of UCS using artificial neural network (ANN), hybrid ANN with imperialism competitive algorithm (ICA–ANN), hybrid ANN with artificial bee colony (ABC–ANN) and genetic programming (GP) approaches. To reach the aim of the current study, an experimental database containing a total of 71 data sets was set up by performing a number of laboratory tests on the rock samples collected from a tunnel site in Malaysia. To construct the desired predictive models of UCS based on training and test patterns, a combination of several rock characteristics with the most influence on UCS has been used as input parameters, i.e. porosity (n), Schmidt hammer rebound number (R), p-wave velocity (Vp) and point load strength index (Is(50)). To evaluate and compare the prediction precision of the developed models, a series of statistical indices, such as root mean squared error (RMSE), determination coefficient (R2) and variance account for (VAF) are utilized. Based on the simulation results and the measured indices, it was observed that the proposed GP model with the training and test RMSE values 0.0726 and 0.0691, respectively, gives better performance as compared to the other proposed models with values of (0.0740 and 0.0885), (0.0785 and 0.0742), and (0.0746 and 0.0771) for ANN, ICA–ANN and ABC–ANN, respectively. Moreover, a parametric analysis is accomplished on the proposed GP model to further verify its generalization capability. Hence, this GP-based model can be considered as a new applicable equation to accurately estimate the uniaxial compressive strength of granite block samples.

The successful use of fly ash (FA) and silica fume (SF) materials has been reported in the design of concrete samples in the literature. Due to the benefits of using these materials, they can be utilized in many industrial applications. However, the proper use of them in the right mixes is one of the important factors with respect to the strength and weight of concrete. Therefore, this paper develops relationships based on meta-heuristic (MH) algorithms (artificial bee colony technique) to evaluate the compressive strength of concrete specimens using laboratory experiments. A database comprising silica fume replacement ratio, fly ash replacement ratio, total cementitious material, water content coarse aggregate, high-rate water-reducing agent, fine aggregate, and age of samples, as model inputs, was used to evaluate and predict the compressive strength of concrete samples. Developed models of the MH technique created relationships between the mentioned parameters. In the new models, the influence of each parameter on the compressive strength was determined. Finally, using the developed model, optimum conditions for compressive strength of concrete samples were presented. This paper demonstrated that the MH algorithms are able to develop relationships that can serve as good substitutes for empirical models.

Prediction of ultimate pile bearing capacity with the aid of field experimental results through artificial intelligence (AI) techniques is one of the most significant and complicated problem in pile analysis and design. The aim of this research is to develop a new AI predictive models for predicting pile bearing capacity. The first predictive model was developed based on the combination of adaptive neuro-fuzzy inference system (ANFIS) and group method of data handling (GMDH) structure optimized by particle swarm optimization (PSO) algorithm called as ANFIS–GMDH–PSO model; the second model introduced as fuzzy polynomial neural network type group method of data handling (FPNN–GMDH) model. A database consists of different piles property and soil characteristics, collected from literature including CPT and pile loading test results which applied for training and testing process of developed models. Also a common artificial neural network (ANN) model was applied as a reference model for comparing and verifying among hybrid developed models for prediction. The modelling results indicated that improved ANFIS–GMDH model achieved relatively higher performance compared to ANN and FPNN–GMDH models in terms of accuracy and reliability level based on standard statistical performance indices such as coefficient of correlation (R), mean square error, root mean square error and error standard deviation values.

Fly-rock caused by blasting is one of the dangerous side effects that need to be accurately predicted in open-pit mines. This study proposed a new technique to predict the distance of fly-rock based on an ensemble of support vector regression models (SVRs) and Lasso and elastic-net regularized generalized linear model (GLMNET), called SVRs–GLMNET. It was developed based on a combination of six SVR models and a GLMNET model. Accordingly, the dataset including 210 experimental data was divided into three parts, i.e., training, validating, and testing. Of the whole dataset, 70% was used for the development of the six SVR models first as the sub-models. Subsequently, 20% of the entire dataset (the validating dataset) was used to predict fly-rock based on the six developed SVR models. The predicted results from the six developed SVR models were used as the input variables to establish the GLMNET model (i.e., SVRs–GLMNET model). Finally, the remaining 10% of the dataset was used for testing the performance of the proposed SVRs–GLMNET model. A comparison and evaluation of the six developed SVR models and the proposed SVRs–GLMNET model were implemented based on five statistical criteria, such as mean absolute error (MAE), mean absolute percentage error (MAPE), root-mean-square error (RMSE), variance account for (VAF), and determination of correlation (R2). The results indicated that the proposed SVRs–GLMNET model provided the most dominant performance in predicting the distance of fly-rock caused by bench blasting in this study with an RMSE of 3.737, R2 of 0.993, MAE of 3.214, MAPE of 0.018, and VAF of 99.207. Whereas, the other models yielded poorer accuracy with RMSE of 7.058–12.779, R2 of 0.920–0.972, MAE of 3.438–7.848, MAPE of 0.021–0.055, and VAF of 90.538–97.003.

A wide variety of artificial intelligence methods have been utilized in the prediction of flyrock induced by blasting. This study focuses on developing a model based on deep neural network (DNN) which is an advanced version of artificial neural network (ANN) for the prediction of flyrock based on the data obtained from the Ulu Thiram quarry that is located in Malaysia. To evaluate and document the success and reliability of the new DNN model, an ANN model based on five different data categories from the established database, was also developed and then compared with the DNN model. Based on the obtained results [i.e. coefficient of determination (R2) = 0.9829 and 0.9781, root mean square error (RMSE) = 8.2690 and 9.1119 for DNN and R2 = 0.9093 and 0.8539, RMSE = 19.0795 and 25.05120 for ANN], a significant increase in predicting flyrock is achieved by developing this DNN predictive model. Then, the DNN model was selected as a function for optimizing flyrock by a powerful optimization technique namely whale optimization algorithm (WOA). The WOA was able to minimize the flyrock resulting from blasting and provide a suitable pattern for blasting operations in mines.

The type of materials used in designing and constructing structures significantly affects the way the structures behave. The performance of concrete and steel, which are used as a composite in columns, has a considerable effect upon the structure behavior under different loading conditions. In this paper, several advanced methods were applied and developed to predict the bearing capacity of the concrete-filled steel tube (CFST) columns in two phases of prediction and optimization. In the prediction phase, bearing capacity values of CFST columns were estimated through developing gene expression programming (GEP)-based tree equation; then, the results were compared with the results obtained from a hybrid model of artificial neural network (ANN) and particle swarm optimization (PSO). In the modeling process, the outer diameter, concrete compressive strength, tensile yield stress of the steel column, thickness of steel cover, and the length of the samples were considered as the model inputs. After a series of analyses, the best predictive models were selected based on the coefficient of determination (R2) results. R2 values of 0.928 and 0.939 for training and testing datasets of the selected GEP-based tree equation, respectively, demonstrated that GEP was able to provide higher performance capacity compared to PSO–ANN model with R2 values of 0.910 and 0.904 and ANN with R2 values of 0.895 and 0.881. In the optimization phase, whale optimization algorithm (WOA), which has not yet been applied in structural engineering, was selected and developed to maximize the results of the bearing capacity. Based on the obtained results, WOA, by increasing bearing capacity to 23436.63 kN, was able to maximize significantly the bearing capacity of CFST columns.

This paper is aimed to develop and verify a novel hybrid intelligent model, which is indeed a new version of GMDH algorithm and named generalized structure of GMDH (GS-GMDH) to solve engineering problems. The proposed GS-GMDH model was validated its capability of predicting blast-induced ground vibration, a very important safety issues in the mining industry. For this regard, a data set with a totally of 96 samples was gathered from a blasting site in Shur River Dam region, Iran. Among them, 67 and 29 samples were used for constructing and testing the model, respectively. To check the accuracy and robustness of the proposed algorithm, the values of the performance evaluation measures, i.e., R2, variance accounted for (VAF), mean absolute error, root mean square error, scatter index, and Nash–Sutcliffe model efficiency (EN–S), were used. The results showed the efficiency of the GS-GMDH algorithm in the prediction of the blast-induced ground vibration. It was also confirmed that the proposed algorithm can be applied effectively to solving/predicting the engineering problems and it has also the potential to be generalized to other fields.

As blasting has some environmental constraints, ripping has got more prevalent as a method of breaking ground in both civil and mining engineering applications. A rippability model of a higher applicability is, therefore, needed to appropriately predict the ripping production (Q) prior to carrying out such tests. For the purpose of predicting the results of ripping production that were attained in three sites located in Johor state, Malaysia, the present study applied three hybrid intelligent techniques, i.e., neuro-bee, neuro-imperialism, and neuro-swarm, to predict Q. In fact, the effects of artificial bee colony (ABC), imperialism competitive algorithm (ICA), and particle swarm optimization (PSO) on weights and biases of neural networks were examined in the current research to receive better prediction/evaluation of ripping production. To do this, totally, 74 ripping tests were taken into consideration in the investigated regions and their influential parameters were assessed. Many parametric studies on ABC, ICA, and PSO parameters were conducted, and then, a comparison was done on the obtained results of the predictive hybrid models using several performance indices. As confirmed by the comparative results, the neuro-bee model proposed in this study estimated Q with a higher accuracy than the other hybrid models. The root-mean-square error (RMSE) values of 0.060, 0.076, and 0.094 were obtained for testing the data sets of neuro-bee, neuro-imperialism, and neuro-swarm techniques, respectively. This clearly shows that the newly developed hybrid model was superior to its rival in terms of predicting the ripping production. Furthermore, results of sensitivity analysis showed that weathering zone is the most influential factor on ripping production as compared to other inputs.

Prediction of pile-bearing capacity developing artificial intelligence models has been done over the last decade. Such predictive tools can assist geotechnical engineers to easily determine the ultimate pile bearing capacity instead of conducting any difficult field tests. The main aim of this study is to predict the bearing capacity of pile developing several smart models, i.e., neuro-genetic, neuro-imperialism, genetic programing (GP) and artificial neural network (ANN). For this purpose, a number of concrete pile characteristics and its dynamic load test specifications were investigated to select pile cross-sectional area, pile length, pile set, hammer weight and drop height as five input variables which have the most impacts on pile bearing capacity as the single output variable. It should be noted that all the aforementioned parameters were measured by conducting a series of pile driving analyzer tests on precast concrete piles located in Pekanbaru, Indonesia. The recorded data were used to establish a database of 50 test cases. With regard to data modelling, many smart models of neuro-genetic, neuro-imperialism, GP and ANN were developed and then evaluated based on the three most common statistical indices, i.e., root mean squared error (RMSE), coefficient determination (R2) and variance account for (VAF). Based on the simulation results and the computed indices’ values, it is observed that the proposed GP model with training and test RMSE values of 0.041 and 0.040, respectively, performs noticeably better than the proposed neuro-genetic model with RMSE values of 0.042 and 0.040, neuro-imperialism model with RMSE values of 0.045 and 0.059, and ANN model with RMSE values of 0.116 and 0.108 for training and test sets, respectively. Therefore, this GP-based model can provide a new applicable equation to effectively predict the ultimate pile bearing capacity.

The stability of retaining walls against overturning is analyzed in this study using artificial intelligence methods. Five input parameters including wall height, wall thickness, soil friction angle, soil density, and stone cement mixture density were varied and 2000 cases were considered in developing the predictive models. Using the artificial neural network (ANN) method, eight prediction models were developed and evaluated based on the coefficient of determination (R2) and the root mean square error. R2 values of 0.9740 and 0.9824 for training and testing datasets, respectively (for the best model), indicate the level of ANN capability in predicting safety factor (SF) of retaining walls. After developing the ANN model, the ant colony optimization (ACO) algorithm was used to maximize the safety factor of the wall by varying the input parameters. In fact, the best ANN model was selected to be used as a modeling function in ACO algorithm. The SF result from optimization section was obtained as 3.057 which show a significant difference from the mean SF values used in the modeling. It can be concluded that ACO may be used as a powerful optimization algorithm in optimizing SF results of retaining walls.

Prediction of tunnel boring machine (TBM) performance parameters can be caused to reduce the risks associated with tunneling projects. This study is aimed to introduce a new hybrid model namely Firefly algorithm (FA) combined by artificial neural network (ANN) for solving problems in the field of geotechnical engineering particularly for estimation of penetration rate (PR) of TBM. For this purpose, the results obtained from the field observations and laboratory tests were considered as model inputs to estimate PR of TBMs operated in a water transfer tunnel in Malaysia. Five rock mass and material properties (rock strength, tensile strength of rock, rock quality designation, rock mass rating and weathering zone) and two machine factors (trust force and revolution per minute) were used in the new model for predicting PA. FA algorithm was used to optimize weight and bias of ANN to obtain a higher level of accuracy. A series of hybrid FA-ANN models using the most influential parameters on FA were constructed to estimate PR. For comparison, a simple ANN model was built to predict PR of TBM. This ANN model was improved on the basis of new ways. By doing this, the best ANN model was chosen for comparison purposes. After implementing the best models for two methods, the data were divided into five separate categories. This will minimize the chance of randomness. Then the best models were applied for these new categories. The results demonstrated that new hybrid intelligent model is able to provide higher performance capacity for predicting. Based on the coefficient of determination 0.948 and 0.936 and 0.885 and 0.889 for training and testing datasets of FA-ANN and ANN models, respectively, it was found that the new hybrid model can be introduced as a superior model for solving geotechnical engineering problems.

When working on underground projects, especially where ground is burst prone, it is of a high significance to accurately predict the risk of rockburst. The present paper integrates the firefly algorithm (FA) and artificial neural network (ANN) aiming at modeling the complex relationship between the rockburst risk in deep mines and tunnels and factors effective on this phenomenon. The model was established and validated through the use of a data set extracted from previously conducted studies. The data set involves a total of 196 reliable rockburst cases. The use of smart systems was used to classify and determine patterns in this research using model development. The hybrid FA–ANN model provides a solution for determining different classes of hazard under different conditions. The capability of these developed systems was implemented to determine the four types of levels defined for this phenomenon. The results of these systems led to new solutions to classify this phenomenon by success rates. Each system, given its performance, yields a unique error. Finally, by combining the number of correctly classified classes and their error values, the success rates in the classification of rockburst phenomena in mines and underground tunnels were evaluated.

The aim of this research is to develop three soft-computing techniques, including adaptive-neuro-fuzzy inference system (ANFIS), genetic-programming (GP) tree-based, and simulated annealing–GP or SA–GP for prediction of the ultimate-bearing capacity (Qult) of the pile. The collected database consists of 50 driven piles properties with pile length, pile cross-sectional area, hammer weight, pile set and drop height as model inputs and Qult as model output. Many GP and SA–GP models were constructed for estimating pile bearing capacity and the best models were selected using some performance indices. For comparison purposes, the ANFIS model was also applied to predict Qult of the pile. It was observed that the developed models are able to provide higher prediction performance in the design of Qult of the pile. Concerning the coefficient of correlation, and mean square error, the SA–GP model had the best values for both training and testing data sets, followed by the GP and ANFIS models, respectively. It implies that the neural-based predictive machine learning techniques like ANFIS are not as powerful as evolutionary predictive machine learning techniques like GP and SA–GP in estimating the ultimate-bearing capacity of the pile. Besides, GP and SA–GP can propose a formula for Qult prediction which is a privilege of these models over the ANFIS predictive model. The sensitivity analysis also showed that the Qult of pile looks to be more affected by pile cross-sectional area and pile set.

Flyrock is an adverse effect produced by blasting in open-pit mines and tunnelling projects. So, it seems that the precise estimation of flyrock is essential in minimizing environmental effects induced by blasting. In this study, an attempt has been made to evaluate/predict flyrock induced by blasting through applying three hybrid intelligent systems, namely imperialist competitive algorithm (ICA)–artificial neural network (ANN), genetic algorithm (GA)–ANN and particle swarm optimization (PSO)–ANN. In fact, ICA, PSO and GA were used to adjust weights and biases of ANN model. To achieve the aim of this study, a database composed of 262 datasets with six model inputs including burden to spacing ratio, blast-hole diameter, powder factor, stemming length, the maximum charge per delay, and blast-hole depth and one output (flyrock distance) was established. Several parametric investigations were conducted to determine the most effective factors of GA, ICA and PSO algorithms. Then, at the end of modelling process of each hybrid model, eight models were constructed and their results were checked considering two performance indices, i.e., root mean square error (RMSE) and coefficient of determination (R2). The obtained results showed that although all predictive models are able to approximate flyrock, PSO–ANN predictive model can perform better compared to others. Based on R2, values of (0.943, 0.958 and 0.930) and (0.958, 0.959 and 0.932) were found for training and testing of ICA–ANN, PSO–ANN and GA–ANN predictive models, respectively. In addition, RMSE values of (0.052, 0.045 and 0.057) and (0.045, 0.044 and 0.058) were achieved for training and testing of ICA–ANN, PSO–ANN and GA–ANN predictive models, respectively. These results show higher efficiency of the PSO–ANN model in predicting flyrock distance resulting from blasting. Moreover, sensitivity analysis shows that hole diameter is more effective than others.