It was 35 years ago since the first usage of Artificial Intelligence (AI) technique in geotechnical engineering, during those years many (AI) techniques were developed based in mathematical, statistical and logical concepts, but the breakthrough occurs by mimicking the natural searching and optimization algorithms. This huge development in (AI) techniques reflected on the geotechnical engineering problems. In this research, 626 paper and thesis published in the period from 1984 to 2019 concerned in applying (AI) techniques in geotechnical engineering were collected, filtered, arranged and classified with respect to subject, (AI) technique, publisher and publishing date and stored in a database. The extracted information from the database were tabulated, presented graphically and commented. The main conclusions is that the number of researches in this field increases almost exponentially, the most used (AI) technique is the Artificial Neural Networks and its enhancements where it is presents about half the researches and finally correlating soil and rock properties is the most addressed subject with about 30% of the researches.

35 Years of (AI) in Geotechnical Engineering: State of the Art

Engineering experience may be written in mathematical form by using adaptive network-based fuzzy inference system (ANFIS). In this article we propose a method to use engineering experience and build a model, which can be used as a systematic decision support tool for engineers dealing with new problems. Planning geotechnical investigations is based on experience, which are used to obtain optimal number of investigation points, field and laboratory tests. To achieve this objective we define minimum number of investigation points and several input parameters which could increase or decrease the number of investigation points. The expert’s evaluations were put in a table, from which we generate the basis of the system. The paper presents a concept for planning geotechnical investigation for buildings using ANFIS and practical examples show its usefulness.

Planning Geotechnical Investigation Using ANFIS

Data-driven determination of sample number and efficient sampling locations for geotechnical site investigation of a cross-section using Voronoi diagram and Bayesian compressive sampling

Unsaturated lateritic soils are complex soils to work with due to moisture effects. So, the determination of its properties requires lots of time, labor and equipment. For this reason, the application of evolutionary learning techniques has been adopted to overcome these complexities. Lateritic soil under unsaturated condition classified as poorly graded and A-7–6 group was subjected to treatment by using hybrid cement and nanostructured quarry fines in a stabilization method. The clay activity, clay content and frictional angle were determined through multiple experiments at different proportions of the additives. 121 datasets were collected through the multiple testing of treated specimens and 70% and 30% of the datasets were used in the model training and testing, respectively to predict the coefficients of curvature and uniformity (Cc and Cu) of the unsaturated lateritic soil. Fist, the multi-linear regression (MLR) model showed that the selected input parameters correlated well with the output parameters. The model performance evaluation and validation selected indicators; R2, RMSE and MAE showed that ANFIS with 0.9999, 0.0021 and 0.0015 respectively, for the training and 0.9994, 0.0077 and 0.0059 respectively outclassed all its hybrid techniques and MLR in both training and testing. However, ANFIS-PSO with performance indicators 0.9996, 0.0062 and 0.0050 respectively (training) and 0.9989, 0.0095 and 0.0073 respectively (testing); followed by ANFIS-GA; 0.9991, 0.0094, and 0.0065 respectively (training) and 0.0089, 0.0099, and 0.0079 (testing) outclassed the other learning techniques for the Cc prediction model while ANFIS-GA; 0.9949, 0.1000, and 0.0798 respectively (training) and 0.9954, 0.0983, and 0.0807 respectively, followed by ANFIS-PSO; 0.9893, 0.1347, and 0.1011 respectively (training) and 0.9951, 0.1127, and 0.0924 respectively outclassed the other techniques for the Cu prediction model. Finally, ANFIS and its evolutionary hybrid techniques have shown their usefulness and flexibility in predicting stabilized unsaturated soil properties for sustainable earthwork design, construction and foundation performance monitoring.

Application of ANFIS hybrids to predict coefficients of curvature and uniformity of treated unsaturated lateritic soil for sustainable earthworks

A fuzzy sets decision support system is proposed for geotechnical site investigation. The system considers parameters such as geology and soil variability that affect the required number of soundings to adequately characterize a site. It permits also to consider vagueness and lack of information. On the basis of available qualitative and quantitative information, the system allows estimating, for common projects, the number of site soundings. Monte Carlo simulations of entry ranges, where each point has a uniform probability distribution, permit to arrange the output in form of histograms fitted with probability functions. The cases presented show that the fuzzy inference system can be used as a systematic decision support for engineers dealing with site characterization.

Site Soundings Density for Geotechnical Investigation with Combined Fuzzy and Probabilistic Input Information

A geotechnical site investigation is an important and complex task that is generally carried out in two steps. The first step, consisting of preliminary soundings, guides the subsequent site characterization. The number of soundings required to adequately characterize a site is set on the basis of an engineering judgement following the preliminary investigation, this is affected by the geological context, the area topography, the project type, and the knowledge of the neighbouring areas. A fuzzy-sets decision-support system, considering parameters that affect the number of soundings required to adequately characterize a site, is proposed. Parameter uncertainties and a lack of information are also considered. On the basis of the available qualitative and quantitative information, the proposed fuzzy system makes it possible to estimate, for a common project, the number of site soundings required to adequately characterize the site. The cases presented show that a Fuzzy Inference System can be used as a systematic decision-support tool for engineers dealing with site characterizations.

Fuzzy-sets decision-support system for geotechnical site soundings

Limit equilibrium methods have been extensively used in the design of reinforced soil structures using various types of failure surfaces. A significant problem in limit equilibrium analysis of reinforced soil is the need to know the force at each reinforcement layer in the limit state. Limit equilibrium analysis with log spiral failure mechanism has been employed to locate the critical failure surfaces emerging at or above the toe and to study the behavior of geosynthetic-reinforced soil slopes. This paper describes a numerical procedure, using limit equilibrium methods, to evaluate the distribution of the soil slope reinforcement tensile forces. Both, log spiral and circular slip surfaces are considered. The effect on the required tensile forces of the target safety factor value, the ratio of the mobilized soil shear strength, and the reinforcement layer spacing are examined. Results show that the distribution of reinforcement tensile forces is dependant on reinforcement layer spacing; however, the total reinforcement force is not. The total reinforcement force required to ensure equilibrium is strongly dependant on the target safety factor value and the ratio of the mobilized soil shear strength. Modified Bishop Method with circular failure surface and Shiwakoti–Leshchinsky log spiral failure surface led to comparable results.

A New Approach for Evaluating the Soil Slope Reinforcement Tensile Forces Using Limit Equilibrium Methods

The importance of the retaining structures is crucial in geotechnical engineering and the accurate determination of earth pressures and natural frequency is important in order to study the dynamic behavior of these structures. The analytical formulae which do not consider the earth pressures behind the retaining structure are usually used. An analytical model for determining the natural frequency of retaining structures including the earth pressures by failure wedges was proposed in the present analysis. The model considers the effect of Coulomb and Mononobe Okabe earth pressures. The current model was validated using several analytical models reported in the literature of the earlier researchers.

Smain Belkacemi

Papers

Fuzzy-sets decision-support system for geotechnical site soundings

Site Soundings Density for Geotechnical Investigation with Combined Fuzzy and Probabilistic Input Information

A New Approach for Evaluating the Soil Slope Reinforcement Tensile Forces Using Limit Equilibrium Methods

An Analytical Model for Determining the Natural Frequency of Retaining Structures Including the Earth Pressures

Diffusion des contraintes dans un milieu granulaire caractérisé par un coefficient de poussée latérale