BULETINUL INSTITUTULUI POLITEHNIC DIN IAŞI Publicat de Universitatea Tehnică „Gheorghe Asachi” din Iaşi Tomul LIX (LXIII), Fasc. 6, 2013 Secţia CONSTRUCŢII. ARHITECTURĂ NEURAL NETWORKS USED IN DESIGN OF REINFORCED LAYER FOR EXISTING SLABS FOR AIRPORT RIGID RUNWAY STRUCTURES BY GABRIELA COVATARIU * “Gheorghe Asachi” Technical University of Iaşi Faculty of Civil Engineering and Building Services Received: November 1, 2013 Accepted for publication: November 19, 2013 Abstract. In this paper a method of using neural networks for improving the computing method by increasing the accuracy in design of the reinforced concrete slabs from airport infrastructure is presented. The obtained results after the models developed with the method of finite element were used in order to create a neural networks simulating the function H R =f (H e , c ss , K, adm ), for dual type of landing gear, for each loading, reaction modulus considered, to design the reinforced layer for existing cement concrete slabs. The use of neural networks for the interpolations of functions to dimension the slabs proved an increase of result accuracy compared to the reading of nomograms, previously carried out, as well as the possibility of computing the variable concrete slab thickness, other than the one considered for the nomograms. Key words: neural networks; airport reinforced slabs; nomograms; runway structural design. 1. Introduction The evolution of air traffic (intensity, types of aircraft) requires dimensioning of airport surfaces, with a high level of confidence. This airports * Corresponding author: covagab@ce.tuiasi.ro