Achieving good results in applying artificial neural networks (ANN) in predicting requires some preparatory works on the set of data. One of them is standardization which is necessary when nonlinear activation function is applied. Basing on predicting completion period of building contracts by multi-layer ANN with error backpropagation algorithm, six different methods of input data standardization were checked in order to determine which allows to achieve the most accurate predictions

The problem of designing deep foundations is related to many civil engineering structures as it is becoming more common and frequent to construct buildings on soft soils. Pile foundation is a popular deep foundation type used to transfer superstructure load into subsoil and bearing layers. However, accurate prediction of piles’ settlement is particularly difficult concerning complicated consolidation process and pile-soil interaction. The objective of this paper is to model a soil-pile system using FEM implemented in Abaqus software. The numerical results of pile bearing capacity and pile settlement were compared with static load test results of CFA piles carried out during construction of Łazienkowska tract flyovers in Warsaw and with engineering analytical calculations according to Eurocode 7 and Polish Standard Code

The present paper is dedicated to the analysis of under sleeper pads (USP), which are resilient elements used in ballasted track systems as vibration isolators. Four types of USP are considered. The authors present the results of laboratory tests, which are then used as input values for the finite element (FE) and mechanical model of the structure. A special focus is put on the description of an original four-degree-of-freedom (4DoF) mechanical model of the system that includes a fractional rheological model of USP. Using the proposed approaches, the dynamic characteristics of under sleeper pads are determined, and conclusions on vibration isolation effectiveness are drawn.