Optimal Prediction of Time-to-Failure from Information Revealed by Damage

Details Optimal Prediction of Time-to-Failure from Information Revealed by Damage Optimal Prediction of Time-to-Failure from Information Revealed by Damage

2005-11-05

arxiv.org/abs/cond-mat/0511134v1

Europhysics. Letters 74 (5), 778-784 (2006)

Physics

Condensed Matter

Materials Science

7 pages with 2 figures

Scientific paper

10.1209/epl/i2006-10036-6

We present a general prediction scheme of failure times based on updating continuously with time the probability for failure of the global system, conditioned on the information revealed on the pre-existing idiosyncratic realization of the system by the damage that has occurred until the present time. Its implementation on a simple prototype system of interacting elements with unknown random lifetimes undergoing irreversible damage until a global rupture occurs shows that the most probable predicted failure time (mode) may evolve non-monotonically with time as information is incorporated in the prediction scheme. In addition, both the mode, its standard deviation and, in fact, the full distribution of predicted failure times exhibit sensitive dependence on the realization of the system, similarly to ``chaos'' in spinglasses, providing a multi-dimensional dynamical explanation for the broad distribution of failure times observed in many empirical situations.

Affiliated with

Also associated with

No associations

Rating

Optimal Prediction of Time-to-Failure from Information Revealed by Damage does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Optimal Prediction of Time-to-Failure from Information Revealed by Damage, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Optimal Prediction of Time-to-Failure from Information Revealed by Damage will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-419366