Minimisation of the residual stress in the heat affected zone by means of numerical methods
Materials
& Design
Materials and Design 28 (2007) 2295–2302 www.elsevier.com/locate/matdes Minimisation of the residual stress in the heat affected zone by means of numerical methods
A.G. Olabi b a,*
, G. Casalino b, K.Y. Benyounis c, A. Rotondo
b
a
School of Mechanical and Manufacturing Engineering, Dublin City University, Dublin 9, Ireland
Politecnico di Bari, Dipartimento di Ingegneria Meccanica e Gestionale, Viale Japigia 182, 70126, Italy c Industrial Engineering Department, Garyounis University, Benghazi, Libya
Received 26 January 2006; accepted 4 August 2006
Available online 24 October 2006
Abstract
Minimisation of the residual stress in the heat affected zones of AISI 304 plates was performed by means of several design of the experiments (DOE) techniques. A common problem in post trials DOE analysis is the lack of data. Before planning additional experimental trials in order to complete a design matrix, trying to estimate missing data by means of prediction models could be an interesting alternative. In this paper, the effectiveness of a polynomial regression model as a model for evaluating missed data from experimental trials, was analysed. Results coming from the different experimental designs were compared. The reliability of the solutions was limited to the ranges of the investigated parameters. Extrapolation over those bounds would limit their applicability.
Ó 2006 Elsevier Ltd. All rights reserved.
Keywords: DOE; Missing data; Regression; RSM; Factorial design; Taguchi method of DOE
1. Introduction
Continuously pursuing high quality and low cost in every kind of industrial processes has made statistical analysis a common and fundamental practice in several business environments. In particular, optimization techniques, related or not with design of the experiments (DOE), support the definition of factors that influence the final output as well as the search for the optimal combinations of the parameters levels.
& Design
Materials and Design 28 (2007) 2295–2302 www.elsevier.com/locate/matdes Minimisation of the residual stress in the heat affected zone by means of numerical methods
A.G. Olabi b a,*
, G. Casalino b, K.Y. Benyounis c, A. Rotondo
b
a
School of Mechanical and Manufacturing Engineering, Dublin City University, Dublin 9, Ireland
Politecnico di Bari, Dipartimento di Ingegneria Meccanica e Gestionale, Viale Japigia 182, 70126, Italy c Industrial Engineering Department, Garyounis University, Benghazi, Libya
Received 26 January 2006; accepted 4 August 2006
Available online 24 October 2006
Abstract
Minimisation of the residual stress in the heat affected zones of AISI 304 plates was performed by means of several design of the experiments (DOE) techniques. A common problem in post trials DOE analysis is the lack of data. Before planning additional experimental trials in order to complete a design matrix, trying to estimate missing data by means of prediction models could be an interesting alternative. In this paper, the effectiveness of a polynomial regression model as a model for evaluating missed data from experimental trials, was analysed. Results coming from the different experimental designs were compared. The reliability of the solutions was limited to the ranges of the investigated parameters. Extrapolation over those bounds would limit their applicability.
Ó 2006 Elsevier Ltd. All rights reserved.
Keywords: DOE; Missing data; Regression; RSM; Factorial design; Taguchi method of DOE
1. Introduction
Continuously pursuing high quality and low cost in every kind of industrial processes has made statistical analysis a common and fundamental practice in several business environments. In particular, optimization techniques, related or not with design of the experiments (DOE), support the definition of factors that influence the final output as well as the search for the optimal combinations of the parameters levels.
References: [6] MINITABTM, Statistical Software Release 14, 2004.