Degradation modeling based on gamma process models with random effects |
| |
| Authors: | Luis Alberto Rodríguez-Picón Anna Patricia Rodríguez-Picón Luis Carlos Méndez-González Manuel I Rodríguez-Borbón Alejandro Alvarado-Iniesta |
| |
| Institution: | 1. Department of Industrial Engineering and Manufacturing, Institute of Engineering and Technology, Autonomous University of Ciudad Juárez, Ciudad Juárez, Chihuahua, México;2. Post Graduate and Research Studies Division, Technological Institute of Ciudad Juarez, Ciudad Juárez, Chihuahua, México |
| |
| Abstract: | The random effects in a gamma process are introduced in terms of its scale parameter. However, the scale parameter affects both its mean and variance. Hence, the variation of the degradation rates and the within degradation increments are expected to be large. For some products, the random effects affect just the rate or just the volatility of the process. Thus, two modifications of the parameters' structure of the gamma process are proposed. One implies that the random effects affect just the volatility and the second just the rate. A Bayesian estimation approach is provided and implemented in two case studies. |
| |
| Keywords: | Bayesian inference Degradation modeling Gamma process Random effects |
|
|
