Some railroad settlement models - a critical review

Mathematical models to simulate railroad track settlements are reviewed and commented upon. There do not seem to be any generally accepted damage and settlement equations describing the long-term behaviour of the track. This also seems to be the case for the ballast material. Most descriptions of the settlement found in the literature are empirical; only different suggestions to describe the track settlement from a phenomenological point of view are available. The track settlement is mostly considered to be a function of number of loading cycles and/or a function of the magnitude of the loading. The settlement should also be a function of the properties of the ballast and subground materials, but very little has been found on this in the literature. By use of the finite element pro gram LS-DYNA, a computer model (very simple) has been created to simulate the long-term behaviour of the track. The model consists of a rail, rigid sleepers, non-linear ballast springs (stiffnesses) and ballast damping, In a solid element beneath each ballast spring, track settlement can be accumulated. Settlement will occur if the stresses in that element exceed a yield limit of the element material. Also 'hanging sleepers' may be modelled and obtained as a result of the track settlement.