Liquefaction study of fine-grained soil using computational model

Liquefaction is one of the most disastrous phenomena that arises due to earthquakes and has always been a major concern for engineers due to the damages and devastation it causes to the environment, structures and the human life. Liquefaction evaluation has been studied vigorously by many researchers for past few decades and based on their observations various researchers gave different limits of PI and other geotechnical parameters which classified soil in liquefiable, potentially liquefiable and non-liquefiable zones, but the question of reliability still needs to be addressed. The present study provides a new set of range for plasticity index and wc/LL ratio for liquefaction classification of fine-grained soil. The present study develops a computational model based on in situ soil properties to evaluate liquefaction potential. Artificial neural network (ANN) model has been developed for predicting liquefaction susceptibility. The significance of plasticity index on liquefaction has been primarily considered while developing the ANN model. The results confirm that the use of artificial intelligence shows the best success rate amongst all the considered approaches for prediction of liquefaction. Due to its efficient cost and quick predictions, it can be used as a sustainable method for evaluating and predicting risk against seismic hazard and infrastructural development.