Statistics and Probability Letters 83 (2013) 1364–1371 Contents lists available at SciVerse ScienceDirect Statistics and Probability Letters journal homepage: www.elsevier.com/locate/stapro Reliability analysis using ageing intensity function Subarna Bhattacharjee a , Asok K. Nanda b, , Satya Kr. Misra c a Department of Mathematics, Ravenshaw University, Cuttack-753003, Odisha, India b Department of Mathematics and Statistics, IISER Kolkata, Mohanpur Campus, Mohanpur-741252, West Bengal, India c School of Applied Sciences, KIIT University, Bhubaneswar-751024, Odisha, India article info Article history: Received 24 June 2012 Received in revised form 17 December 2012 Accepted 12 January 2013 Available online 4 February 2013 MSC: primary 60E15 secondary 62N05 60E05 Keywords: Ageing phenomenon Generalized Weibull model Hazard rate abstract Ageing intensity (AI) function analyzes the ageing property of a system quantitatively. We study the behavior of a few generalized Weibull models and some system properties in terms of AI function. We establish that AI function provides a major and altogether a new role in studying system’s ageing behavior from reliability perspective. © 2013 Elsevier B.V. All rights reserved. 1. Introduction An important phenomenon in reliability theory is ‘ageing’ which is an inherent property of a unit that may be a living organism or a system of components. Failure rate, mean residual life, etc. are various measures of ageing. Jiang et al. (2003) claim that the representation of ageing of a system by failure rate is qualitative. They developed a new notion, called ageing intensity (AI), to quantitatively evaluate the ageing property of a system, denoted by L X (t ), of a random variable X , defined as the ratio of the instantaneous failure rate r X (t ) to the failure rate average H X (t ) = t 0 r X (u)du /t , for t > 0, i.e., L X (t ) = r X (t ) H X (t ) , where defined, = tf X (t ) ¯ F X (t ) ln ¯ F X (t ) , where f X (·) and ¯ F X (·) are the probability density function and the survival function of the random variable X respectively. Let the random variable X have support S X = (l X , u X ), where u X may be infinity. We define L X (t ) = 0, if t < l X r X (t ) H X (t ) , if l X t u X , if t > u X . Corresponding author. E-mail addresses: asoknanda@yahoo.com, asok@iiserkol.ac.in (A.K. Nanda). 0167-7152/$ – see front matter © 2013 Elsevier B.V. All rights reserved. doi:10.1016/j.spl.2013.01.016