The analysis of failure data, collected directly from the field during the operation of a system, is a well estab- lished research practice to study the dependability behavior of computer-based systems. These studies aim to deeply investigate failure phenomena in existing systems in order to enrich engineers awareness of real issues and improve
successive generations of systems. Traditionally, these stud- ies have focused on the dependability analysis of operating systems, such as, Windows 2000 [10], and Linux [11]. Other studies focused on the analysis of failures of networked systems and more recently, large-scale supercomputers [12]. Recent efforts focused on field failure data analysis of mobile systems as well. An architecture for gathering and analyzing failure data for distributed systems using the Blue- tooth communication protocol is proposed in [13]. In [14] it is discussed the analysis of failure data collected from the base stations of a cellular system, whereas [4] focuses on the analysis of the software bugs collected from Symbian and Android bug reports. None of these works focus on failure data collected directly from mobile phones during their operation. The previous work on Symbian [5], [6] is considered as one of the first examples of a logging system for direct dependability analysis of mobile phones [4]. In [7] we extended our approach to the Android platform, by proposing the design and implementation of a logger for collecting failure data from running Android devices.