Lifetime Analysis of Circular $k$-out-of-$n$: G Balanced Systems in a Shock Environment

TL;DR

This paper analyzes the lifetime distributions of circular $k$-out-of-$n$: G balanced systems in shock environments, modeling their failure times with phase-type distributions using Markov chains and providing computational methods for reliability assessment.

Contribution

It introduces a unified framework for discrete and continuous-time analysis of such systems, demonstrating phase-type distributions for failure times and developing efficient computational techniques.

Findings

SNTF follows a discrete phase-type distribution.

TTF follows a phase-type distribution with different parameters.

Numerical results show the impact of system parameters on reliability metrics.

Abstract

This paper examines the lifetime distributions of circular $k$-out-of-$n$: G balanced systems operating in a shock environment, providing a unified framework for both discrete- and continuous-time perspectives. The system remains functioning only if at least $k$ operating units satisfy a predefined balance condition (BC). Building on this concept, we demonstrate that the shock numbers to failure (SNTF) follow a discrete phase-type distribution by modeling the system's stochastic dynamics with a finite Markov chain and applying BC-based state space consolidation. Additionally, we develop a computationally efficient method for directly computing multi-step transition probabilities of the underlying Markov chain. Next, assuming the inter-arrival times between shocks follow a phase-type distribution, we establish that the continuous-time system lifetime, or the time to system failure (TTF), also follows a phase-type distribution with different parameters. Extensive numerical studies illustrate the impact of key parameters-such as the number of units, minimum requirement of the number of operating units, individual unit reliability, choice of balance condition, and inter-shock time distribution-on the SNTF, TTF, and their variability.