Area: Land management, urbanism, architecture and civil engineering ―Selected Proceedings from the 13th International Congress on Project Engineering‖. (Badajoz, July 2009) APPLICATION OF RESPONSE SURFACE METHODOLOGY FOR THE OPTIMIZATION OF WELDING PARAMETERS AS A FUNCTION OF THE RESULTING THERMAL DISTRIBUTION. Eusebio Martínez Escuela de Arquitectura e Ingeniería de Edificación de UPCT Manuel Estrems Departamento de Materiales e Ingeniería de Fabricación de UPCT Valentín Miguel Escuela de Ingenieros Industriales de Albacete de UCLM Antonio Garrido Escuela de Arquitectura e Ingeniería de Edificación de UPCT José Antonio Guillén Escuela de Arquitectura e Ingeniería de Edificación de UPCT Abstract One of the largest difficulties traditionally found in stainless steel constructions has been the execution of welding parts in them. Nowadays, the technology available permits the use of arc welding processes for that application with certain advantage. Response surface methodology is used to optimize a process in which the variables that take part in it are not related to each other by a mathematical law. Therefore, an empiric model must be formulated. With this methodology the optimization of one selected variable may be done. In the case of welding, it allows to determine the best operation conditions and to minimize the probability of the joint having defects. In this research, an optimization of TIG welding parameters is made as a function of temperature distribution in to the joint: current, voltage, arc efficiency and welding speed. Keywords: Response surface methodology, welding, thermal modelling, statistical design, ANOVA 1. Introduction The response surface methodology (RSM) is a collection of techniques that allow the researcher to inspect a response that can be displayed as a surface, when the experiments investigate the effect of varying quantitative factors in the values that a dependent variable or response takes (Box et., 1951).The RSM is a set of mathematical and statistical techniques that are useful for modeling and analysis in applications where a response of interest is influenced by different variables and the objective is to optimize this response (Garcia-Diaz, 1995). One of the difficulties encountered in building stainless steel structures has been the joints of structural profiles. At present, welding joints is applicable due to the advantages inherent in this procedure. However, the temperatures reached are high because the process involves joining with a fusion of material, which can modify the mechanical, chemical and aesthetic properties of 84