Journal of Advances in Science and Engineering 4 (2021) 86 – 97 Journal of Advances in Science and Engineering (JASE) J o u r n a l h o m e p a g e : www.sciengtexopen.org/index.php/jase Prediction of tungsten inert gas welding process parameter using design of experiment and fuzzy logic P. Pondi  , J. Achebo, A. Ozigagun Department of Production Engineering, Faculty of Engineering, University of Benin, Benin City, Nigeria 1. Introduction Although, welding process parameters are often chosen on the basis of expert judgment or from a welding handbook, this choice does not guaranty the best or near best weld bead profile for that particular welding environment [1]. The quality of a weld depends on several factors including the mechanical properties and reduced post weld defects of the weld, which depends on the chemical composition and metallurgical characteristics of the weld metal [2]. The metallurgical and mechanical properties of a weld depend on the bead geometry, which is directly related to welding process ———  Corresponding author E-mail address: eddyinnaija@gmail.com https://doi.org/10.37121/jase.v4i2.149 2636–607X /2736–0652 © 2021 the authors. Published by Sciengtex. This is an open access article under CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) parameters [3]. It is pertinent to note that post weld defects such as cracks are generated on the weld line when the weld product is subjected to either a bending stress or shocks [4]. However, metallurgical variations linked with fusion welding processes such as solidification cracking, segregation, presence of porosities and grain growth in the heat affected area frequently result in poor mechanical properties of the weldment [5]. Tungsten inert gas (TIG) welding is a process of welding, which involves an arc between non-consumable tungsten electrode and the work piece. The arc, electrode and molten pool are all shielded ARTICLE INFO ABSTRACT Article history: Received 23 February 2021 Received in revised form 25 March 2021 Accepted 29 March 2021 Available online 7 April 2021 The focus of this study is to predict tungsten inert gas (TIG) welding process parameter such as heat input for stabilizing heat and removing post weld crack formation in mild steel weldment. The main input parameters examined are the welding current, voltage and speed whereas the measured (response) parameter is heat input. Statistical design of experiment was done by means of central composite design method using the range and levels of independent variables. The experiment was carried out 20 times (with 5 specimens per run) using 60 mm x 40 mm x 10 mm mild steel coupons. The plate samples were cut longitudinally with a Single-V joint preparation, with the edges beveled. The welding process utilizes 100% pure argon as a protecting gas to shield the weld specimen from external interaction. The interaction between the input and response variables was analyzed using a fuzzy logic system. The result showed that for a welding current, voltage and speed of 190 A, 21 V, and 2.0 mm/s respectively, the predicted heat input was 0.912 kJ/mm whereas for input parameters of (170 A, 25 V, and 2.0 mm/s) and (180 A, 23 V, 0.98 mm/s), the predicted heat inputs were 1.07 kJ/mm and 1.380 kJ/mm, respectively. Keywords: Fuzzy logic Heat input Statistical design of experiment Tungsten inert gas welding