Resistance value (R-value) is one of the basic subgrade stiffness characterizations that express a material’s resistance to deformation. In this paper, artificial intelligence (AI)-based models—especially M5P, support vector machine (SVM), and Gaussian process regression (GPR) algorithms—are built for R-value evaluation that meets the high precision and rapidity requirements in highway engineering. The dataset of this study comprises seven parameters: hydrated lime-activated rice husk ash, liquid limit, plastic limit, plasticity index, optimum moisture content, and maximum dry density. The available data are divided into three parts: training set (70%), test set (15%), and validation set (15%). The output (i.e., R-value) of the developed models is evaluated using the performance measures coefficient of determination (R²), mean absolute error (MAE), relative squared error (RSE), root mean square error (RMSE), relative root mean square error (RRMSE), performance indicator (ρ), and visual framework (Taylor diagram). GPR is concluded to be the best performing model (R², MAE, RSE, RMSE, RRMSE, and ρ equal to 0.9996, 0.0258, 0.0032, 0.0012, 0.0012, and 0.0006, respectively, in the validation phase), very closely followed by SVM, and M5P. The application used for the aforementioned approaches for predicting the R-value is also compared with the recently developed artificial neural network model in the literature. The analysis of performance measures for the R-value dataset demonstrates that all the AI-based models achieved comparatively better and reliable results and thus should be encouraged in further research. Sensitivity analysis suggests that all the input parameters have a significant influence on the output, with maximum dry density being the highest.

Riverbank erosion is a natural process of removal of earthen materials from the bank surface. The process of riverbank erosion that is induced naturally results in the formation of landforms such as valleys, canyons, and productive floodplains. However, riverbank erosion can also be considered a hazard when the process occurs at an alarming rate causing loss of land. The extent of erosion depends on many factors. One of the main factors responsible for riverbank erosion is the soil erodibility which is the resistance of soil to erosion. The aim of this study is to quantify the riverbank erosion rates and the potential magnitude of riverbank erosion in order to generate an empirical predictive model to estimate riverbank erosion from physical and geomorphic variables for rivers susceptible to riverbank erosion. Several models were trained using the Regression Learner application in MATLAB software. Models that include soil erodibility parameters perform better than the models without the soil erodibility parameters. The model with the highest accuracy was found to be Model 2, with Root Mean Square Error (RMSE) of 3.70E-08 and coefficient of determination, R2 of 0.55. The model produced in this study will be helpful to analyze and predict the effects of riverbank erosion and assist in the development of bank stabilization solution. ABSTRAK: Hakisan tebing sungai adalah proses semula jadi terhadap penyingkiran bahan tanah dari permukaan tebing. Proses hakisan tebing sungai yang terjadi secara semula jadi ini mengakibatkan pembentukan bentuk muka bumi seperti lembah, ngarai dan dataran banjir yang produktif. Bagaimanapun, hakisan tebing sungai juga boleh dianggap sebagai ancaman apabila proses berlaku pada kadar membimbangkan sehingga menyebabkan kehilangan tanah. Tahap hakisan bergantung pada banyak faktor. Salah satu faktor utama yang menyebabkan hakisan tebing sungai adalah kebolehhakisan tanah iaitu ketahanan tanah terhadap hakisan. Kajian ini bertujuan untuk mengukur kadar hakisan tebing sungai, mengkaji potensi magnitud hakisan tebing sungai dan menghasilkan model ramalan empirik bagi menganggarkan hakisan tebing sungai daripada pembolehubah fizikal dan geomorfik bagi sungai yang terdedah kepada hakisan tebing sungai. Beberapa model telah dilatih menggunakan aplikasi Regression Learner dalam perisian MATLAB. Dapatan menunjukkan model yang mengandungi parameter kebolehhakisan tanah adalah lebih baik berbanding model tanpa parameter kebolehhakisan tanah. Model 2 didapati mempunyai ketepatan tertinggi dengan ralat punca min kuasa dua (RMSE) sebanyak 3.70E-08 dan pekali penentuan, R2 sebanyak 0.55. Model dalam kajian ini dapat membantu dalam analisa berkaitan kesan hakisan tebing sungai dan penyelesaian kepada pembangunan kestabilan tebing.

Erosion from riverbank is most properly represented by the erodibility factor. One of the methods that can be used for in situ soil erodibility testing is the submerged jet test called Jet Erosion Test (JET). In this study, a newly modified version of the JET device namely Jet Erosion Device (JEd) is fabricated, with improved features and design that facilitates testing in the field and the laboratory. Analysis and calibration of the JEd tests were conducted to check on the reliability and performance of the Jet Erosion Device (JEd). Some preliminary results were shown to give some insights on the capabilities of the JEd. An evaluation of the erosion performance index i.e. jet index was performed to characterize the erosion resistance. The estimation of erodibility coefficients were made using the results of jet index obtained.

Estuaries are bodies of water along the coasts that are formed when fresh water from rivers flows into and mixes with salt water from the ocean. The density of seawater is greater than fresh water and it varies with salinity and temperature. Fresh water tends to float on top of seawater because of its lower density. Human-induced activities like the dredging of shipping lanes along the bottom estuarine, the dumping of industrial wastes into the water system, and shoreline development influence estuarine dynamics which include the mixing process. These activities lead to salinity changes and further adversely affect the estuarine ecosystem. The purpose of this research is to verify how salinity-morphology relations change in an estuarine system under various rainfall patterns, more specifically under extremely high rainfall conditions. The experiment that has been conducted studied the salinity-morphology relationships for a variety of rainfall patterns, most particularly for exceptionally high rainfall conditions, using an idealised channel. In the first part of this research, the morphology changes of the mixing between salt water (estuary) and freshwater (river) for different rainfall patterns, had been investigated in laboratory experiments. Fresh water was released from one end of the flume channel and overflowing over the weir at the other end. Meanwhile, salt water was represented by the red dye tracer released slowly through a weir and intruded horizontally to the upstream as a gravity current. In this experiment, an artificial roughened bed section was used as morphology change. The salinity pattern is plotted using Microsoft Excel. The salinity levels were measured at selected stations along the channel/longitudinal (x-axis), and also in transverse (y-axis) and vertical directions (z-axis) within the time duration. The observed salinity profile showed the difference in salinity level between heavy and light rain conditions with morphology effect where during heavy rain, the salinity level will decrease, hence, the existence of an artificial roughened bed section will affect the time taken for the process of mixing between salt water and fresh water.