New algorithm for systems with simultaneous failures

The algorithm computes the structural signature and mean time-to-failure for every state of a monotone binary system.

An algorithm computes the structural signature and the mean time-to-failure (MTTF) for every state of a monotone binary system. This works instead of only calculating it for the state where all components work.

The method makes no assumption on the structure function. It evaluates the function exactly once per state. Computing the MTTF everywhere costs only a factor polynomial in the number of components over computing it once.

The algorithm identifies all minimal path sets as a by-product. It assumes exchangeable component failure times where the survivor count is a continuous-time Markov chain. This setting allows for simultaneous failures.

The per-state signature is a strictly finer decision criterion than the MTTF. Signature dominance at a common failure count implies MTTF dominance, but not conversely.

Experiments used 22 benchmark networks with up to 30 components and 31 failure models. The best MTTF-threshold policy has a median excess cost of 0.36% over the optimal state-based policy.

Key facts

  • The algorithm computes the MTTF at every state of a monotone binary system.
  • The method works for models that admit simultaneous failures.
  • The best MTTF-threshold policy is never worse than the best policy that repairs after a fixed number of failures.

All sections

Edition