ABSTRACT
This paper presents a life-stress methodology that models the failure rate in the form of a bathtub curve. The model consists of the Alpha Power Transformation (APT), which adds an extra parameter to the probability distributions to achieve better flexibility in the representation in the data analysis. To build the life–stress relationship, the APT is combined with the Weibull Distribution (WD) and the Inverse Power Law (IPL) as a stress model to relate the data from the accelerated life tests (ALT), thus presenting the APTW-IPL. Statistical properties of the APTW-IPL are analyzed and discussed. For the parameter estimation of APTW-IPL, the Maximum Likelihood Estimator was used. On the other hand, to test the efficacy of the APTW-IPL, the model is compared with other methodologies that describe the behavior of the bathtub curve in two case studies related to determining the behavior of electronic devices that were subjected to ALT. The results show that the APTW-IPL can be a good option for reliability analysis in electronic devices. It represents the failure times in the form of a bathtub curve, the value of MTTF, and fitting the distribution to the case study data.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Additional information
Notes on contributors
Luis Carlos Méndez-González
Luis Carlos Méndez González Ph.D. is currently a full-time professor at the Department of Industrial Engineering and Manufacturing at the Autonomous University of Ciudad Juárez, México. He received his Ph.D. in Science in Engineering in 2015. He received his M.S in Industrial Engineering from the Technological Institute of Ciudad Juarez in 2011 and his B.S in Electronics Engineering in 2007. He has more than fifteen years in software, hardware design, Applied Statistics, Measurement System Analysis, Reliability Engineering, and Quality Engineering. He is currently a member of the National Researchers System from México as Level 1. His research interests include reliability and degradation modeling, stochastic modeling, hardware, and software design, and Machine Learning.
Luis Alberto Rodríguez-Picón
Luis Alberto Rodríguez-Picón Ph.D. is currently a full-time professor at the Department of Industrial Engineering and Manufacturing at the Autonomous University of Ciudad Juárez, México. He received his Ph.D. in Science in Engineering in 2015. He received his B.S. and M.S. degrees in Industrial Engineering from the Technological Institute of Ciudad Juárez, México, in 2010 and 2012, respectively. He has worked as a professor in industrial engineering, statistics, and mathematics and has several years of professional experience in the automotive industry. He is currently a member of the National Researchers System from México as Level 1. His research interests include reliability and degradation modeling, stochastic modeling, multivariate statistical modeling, and design of experiments.
Ivan Juan Carlos Pérez-Olguin
Iván JC Pérez-Olguín received a Doctor of Science degree in industrial engineering from the Technological Institute of Ciudad Juarez, Mexico. He is currently a full-time Professor and a Researcher with the Autonomous University of Ciudad Juarez, Mexico. He has published in journals, conference proceedings, and books more than 50 articles; he also contributed to the automotive industry with two patents and four utility models. His research interests include robust optimization, reliability tests, product optimization, process optimization, and lean manufacturing.
Luis Asunción Pérez-Domínguez
Luis Asunción Pérez-Dominguez completed a B.Sc. in Industrial Engineering at Instituto Tecnológico de Villahermosa, Tabasco, México in 2000 and M.Sc. degrees in Industrial Engineering from Instituto Tecnológico de Ciudad Juárez, Chihuahua, México, in 2003 respectively. PhD. Science of Engineering, at the Autonomous University of Ciudad Juárez, Chihuahua, México in 2016. Dr. Luis currently is professor-Research in the Universidad Autónoma de Ciudad Juárez. His research interests include multiple criteria decision-making, fuzzy sets applications and continuous improvement tools applied in the manufacturing field. Member of The Canadian Operational Research Society (CORS); Also, member of Society for Industrial and Applied Mathematics (SIAM). He is recognized as Research associated by Ministerio de Ciencia Tecnología e Innovación, Colombia (Ministry of Science Technology and Innovation in Colombia). He is member of Sistema Nacional de Investigadores recognized by CONACYT, México. Dr. Perez also is a member of EURO Working Group on MCDA (EWG-MCDA).
David Luviano Cruz
David Luviano-Cruz received the Ph.D. degree in sciences from the Centro de Investigación y de Estudios Avanzados del Instituto Politécnico Nacional (CINVESTAV) using artificial neural networks (ANN) to improve path recognition. He is currently an Active Researcher at the Universidad Autónoma de Ciudad Juárez, where he also performs full-time Professor activities. He has published and published more than 23 scientific works with more than 138 citations. His research interests include optimization using artificial neural network algorithms, Pythagorean fuzzy sets, and machine learning.