Prediction and optimization of process parameters using design of experiments and fuzzy logic

This paper presents a mathematical model to investigate the effect of process parameters using response surface methodology, Taguchi technique and fuzzy logic to seek the optimum solution of friction welding parameters. Achievement of the desired quality of any product is possible with the machining and proper selection of set of process parameters. Collected secondary data on Tensile strength and metal loss used for optimizing the process parameters to achieve minimum metal loss and maximum tensile strength. In the present study, three mathematical models were utilized to optimize the weld specimens for Aluminium–silicon carbide particulate (Al. SiCp). Taguchi L orthogonal array used to study the effect of various input friction weld parameters. Mathematical models using response surface methodology (RSM), Taguchi and Fuzzy logic were identified optimum friction welding parameters with less error percentage. Optimum friction weld parameters for determining the minimum metal loss and maximum tensile strength values were carried out in different stages. The R-square values of the different methods for metal loss and tensile strength are presented here. The optimum set of parameters obtained by the Taguchi, Response surface methodology and Fuzzy logic analysis of minimum metal loss and maximum tensile strength values can be compared and provided better results than the observed values. a. Findings: This paper presents simple optimization methodology and its validation with existing test results in finding the range of the output response and optimum parameters of friction weld by modifications in the Taguchi, Fuzzy logic and response surface methodologies. b. Practical implications: Adequacy of this methodology should be examined by considering the test data on friction weld materials and structures with the objective of minimizing the metal loss and maximizing the tensile strength. c. Originality/value: The introduction of fictitious variable and the analysis of ANOVA on experiments of each test run will provide fruitful results and less computation burden with less error percentage. Purpose: The non-linear nature of the output response leads to error with unknown influential input parameters. In addition to this scatter in the outcome of repeated experiments is unavoidable due to various reasons. It is a standard practice to select an orthogonal array in the Taguchi approach for tracing optimum input parameters by conducting a few numbers of experiments and confirm them through additional experimentation (if necessary). The purpose of this paper is to present a simple methodology and its validation with existing test results in finding the range of the output response and optimum parameters of friction weld by modifications in the Taguchi, Fuzzy logic and response surface methodologies. Design/methodology/approach: The modified Taguchi approach, Fuzzy logic and Response surface methodology is proposed to find the optimum process parameters and the expected range of the output response.