29.2 Standardized confidence level

The QC methods described here are based on statistical tests to ensure that directional survey measurements have an accuracy that is compatible with the applicable error model. The testing is based on the
assumption that the measurements have known statistical distributions. In accordance with the recommendations of the ISCWSA, measurements which deviate by more than plus or minus three standard deviations
(3σ) from the expected value are classified as gross errors (noisy measurements); and should be ignored. Only measurements falling within the ±3σ interval should be used to calculate wellbore orientations and positions.

All QC tests and examples presented here are based on the availability of a dedicated error model, the assumption of Normally distributed measurements, and the use of a 3σ test limit is commonly adopted. For a
Gaussian onedimensional error distribution, a 3σ limit corresponds to working at a 99.7% confidence level provided the basic geometrical requirements for the test are fulfilled.