Availability Analysis of a Large-Scale Networked System

• Activity Number: 576
• Type: Contributed
• Date/Time: Wednesday, August 3, 2011 : 2:00 PM to 3:50 PM
• Sponsor: Section on Physical and Engineering Sciences
• Author(s): Shuguang Song*+
• Companies: The Boeing Company
• Keywords: availability ; networked system ; spatio-temporal ; failure time ; repair time ; maintenance

Abstract

In modern industry, complex networked systems have become more and more common. The availability of complex systems is an active research area with application not only in engineering but also in experimental and social sciences. Consider a large-scale networked system with possible heterogeneous N1 sensors, N2 back-up units, i.e. failure time and repair time of different sensors/back-up units could follow different distributions. If a sensor fails at a site, a back-up unit will be transported to the failed sensor site and replace the failed sensor if there is at least one available back-up unit before the failed sensor is repaired back to operating condition. Once the failed sensor is back in operation, then the back-up unit will be immediately replaced by the repaired sensor. Due to the design of the networked system, the physical capability of a sensor/back-up unit, and terrain variations and weather conditions across a large-scale networked system, the "working availability" of the large-scale networked system may not only be time indexed, but also be space indexed. In this paper, we present some results about the availability analysis of such a large-scale networked system.