Reliability analysis of a capacitated series-parallel multi-state system subject to performance sharing mechanism and transmission loss

  • Tianyuan Zhang
  • , Yuchang Mo*
  • , Li Yang
  • *Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

This paper proposes a reliability model of capacitated series-parallel multi-state systems (MSSs) subject to performance sharing and transmission loss. The system consists of m subsystems interconnected through a common bus, where each subsystem contains k parallel-configured components. When a subsystem fails to independently satisfy its demand, the total performance surplus from other subsystems is transmitted to the deficient subsystem through the common bus. The system is deemed unreliable as long as the weighted sum of performance deficiency of subsystems exceeds a pre-designed threshold after performance sharing. The existing methods are computationally intensive due to iterative calculations or system state explosion. A hybrid method combining the continuous-time Markov chain (CTMC) and the multi-valued decision diagram (MDD) is proposed to analyze the system reliability. Specially, the MDD method implements some truncation and merging operations to effectively eliminate redundant branches and compute the reliability of the entire system accurately from the constructed compact MDD model. Case studies of a heating system and a power grid system are provided to illustrate the effectiveness of the proposed method and the practical application of the system model.

Original languageEnglish
Article number111249
JournalReliability Engineering and System Safety
Volume263
DOIs
StatePublished - Nov 2025

Keywords

  • Capacitated series-parallel system
  • Multi-valued decision diagram
  • Performance sharing
  • Reliability analysis
  • Transmission loss

Fingerprint

Dive into the research topics of 'Reliability analysis of a capacitated series-parallel multi-state system subject to performance sharing mechanism and transmission loss'. Together they form a unique fingerprint.

Cite this