Parameter Estimation of Warranty Cost Model using Genetic Algorithm
Nur Hayati Kasim1, Hairudin Abdul Majid2, Azurah A. Samah3
1Nur Hayati Kasim, Department of Modeling and Industrial Computing, Universiti Teknologi Malaysia, Johor Bahru, Malaysia.
2Hairudin Abdul Majid, Department of Modeling and Industrial Computing, Universiti Teknologi Malaysia, Johor Bahru, Malaysia.
3Azurah A. Samah, Department of Modeling and Industrial Computing, Universiti Teknologi Malaysia, Johor Bahru, Malaysia.
Manuscript received on November 01, 2012. | Revised Manuscript received on November 02, 2012. | Manuscript published on November 05, 2012. | PP: 163-166 | Volume-2 Issue-5, November 2012. | Retrieval Number: E1034102512/2012©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Parameter estimation can also be classified as an optimization where the objective is to find the values of some unknown parameters in which the objective function is minimized or maximized. Several methods can be used in estimating parameter including Genetic Algorithm (GA). GA has been widely used in many applications specifically in optimization problem. This paper propose GA as a method to estimate the parameter of warranty cost model with warranty claim data collected from Malaysian automotive industry. Various combinations of GA operators are carried out. Maximum Likelihood Estimation (MLE) method is employed in order to have a comparable solution for the model parameters of warranty cost. The performance of GA in warranty cost model is measured based on Residual Sum of Squares (RSS) and Mean Squared Error (MSE).
Keywords: Genetic Algorithm, parameter estimation, likelihood function, warranty.