Prediction settlements of foundations on soft soils - a case study

Abstract

Excessive settlement of engineering structures and subsequent correction procedure have become a major concern for both the public and the engineering community. The severity of the above problem has increased due to the use of more filled marshy lands for construction. If one studies the problem of excessive settlement of most domestic structures constructed on fills, it could be clearly attributed to the violation of fundamental engineering principles. Lack of engineering knowledge and judgement or sheer negligence of basic engineering principles could be main reasons. If a structure is to be constructed on soft grounds it is clear that the ground should be improved before the construction or the construction procedure should be phased out in such a way that the final settlement, more importantly differential settlement, is within allowable limits for that particular structure. Pre-loading (with or without sand drains), replacement of the soft soil with good material, dynamic replacement and electro-osmosis are some of the methods used for improvement of clayey or peaty deposits. Of the above methods, pre-loading is widely used due to its economy and simplicity. Important considerations in pre-loading are: duration of pre-loading; height of pre-load heap (surcharge load) and the monitoring of settlement experienced. In this paper a case study of construction of a building in marshy lands violating some basic engineering principles is presented. Use of some basic soil properties together with in-situ testing (e.g. Standard Cone Penetration Test) to predict the ultimate settlement under the pre-loading is illustrated. Furthermore, the rate of settlement, an essential component of the pre-loading design, is obtained by curve fitting the settlement data while the pre-loading is in progress. This method of using basic soil properties together with in-situ tests to predict ultimate settlement and curve fitting data to predict the rate of settlement does not require the almost impossible process of obtaining undisturbed soft soil samples for laboratory testing.