Probabilistic Models for Proof Load Testing

Abstract

Proof sampling is the proof load testing of a sample of structural components. A probabilistic model of posterior strength is developed for the untested components in a lot that passes a proof sampling test. When strengths in a lot are highly correlated, it is found that the proof sampling test is virtually equivalent to a proof test of all components. On the other hand, if component strengths are uncorrelated, it is found that there is no point to the test unless all components in the lot are proof tested. Load factors are calculated for the inferred distribution of strength assuming a Gaussian distribution of load. The load factors can be used to find a limit load that the untested components can support with target reliability or, conversely, to specify a required test load for safety verification. In an example concerning lot‐produced roof trusses with moderately high correlation, the proof load testing of two trusses, in situ, avoids other costly testing alternatives.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 11 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.